48
Automatic Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor is Naked! Nuance anti-trust violations continue 3 Speech technology patents issued last month 6 TTS in the call center 3 Nuance accent patent 6 Speech in the Browser 4 Get2Human adds member to Great Customer Service Club: 6 Voice technology in the Supply Chain market 5 Some views on speech 6 Thought for the month: 6 Telephone IVR Applications Products/Services/Partnerships/Financials Atos Origin/Leading Software/Loquendo biometrics partnership 7 Vormetric provides security for Nexidia 11 NH shuts 511 12 Aurix Cloud Based Analysis 8 Nuance NL Fast Track 12 Avaya UC SME apps from partners 8 Nuance/Cyara Solutions partner 12 Dimension Data/Trident Mktg using Cisco CVP 8 Smart Action collects illness symptions 12 Contact Solutions Venture Investment award 8 Syntellect CIM 8.0 13 Convergys expanding call centers 8 Syntellect/Oracle alliance 13 DiRAD at Minnesota 9 CT Connect incentives extended 14 PCI guidance for CC processing 9 Venstar ComfortCall 14 Eckoh Utility payment service 9 Verint adds to Audiolog for contact centers 14 Alcatel-Lucent's Genesys/Enterprise Integrated 10 NG9-1-1 support from Verint 14 RoomPlace Selects Interactive Intelligence 10 Leading Software Selects Voxeo's Cloud 15 KnoahSoft using Aurix 10 VoxGen obtains UK grant 15 Nexidia at BC/BS Tennessee 11 VXi TalkPro Speech Recognition Headsets 15 Telephone Messaging - Applications/Products/ Partnerships/Financial Adeptra recognized by FST magazine 16 Movius expands messaging services 17 Aastra BluStar Video Conferencing phone 16 Nuance Powers Vivo Voicemail-to-Text Service: 18 AVST at ACUTA 16 West launches Campaign Central 18 Jott shuts down 17 Telephone Tools - Applications/Products/ Partnerships/Financial IVONA for Developers is doing well 18 VBVoice provides MS MRCP Connector 19 I6NET/IVONA VoiceXML browser or Asterisk 19 Voxeo Adds Mobile Web Access to Cloud- based VoiceObjects On-Demand 19 Loquendo/NuEcho partner 19

Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

  • Upload
    lehanh

  • View
    216

  • Download
    2

Embed Size (px)

Citation preview

Page 1: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

Automatic Speech Recognition March 2011

Vol. 22 No. 3

INSIDE THIS ISSUE

The Emperor is Naked!

Nuance anti-trust violations continue 3 Speech technology patents issued last month 6

TTS in the call center 3 Nuance accent patent 6

Speech in the Browser 4 Get2Human adds member to Great Customer Service Club:

6 Voice technology in the Supply Chain market 5

Some views on speech 6 Thought for the month: 6

Telephone IVR – Applications Products/Services/Partnerships/Financials

Atos Origin/Leading Software/Loquendo biometrics partnership

7 Vormetric provides security for Nexidia 11

NH shuts 511 12

Aurix Cloud Based Analysis 8 Nuance NL Fast Track 12

Avaya UC SME apps from partners 8 Nuance/Cyara Solutions partner 12

Dimension Data/Trident Mktg using Cisco CVP

8 Smart Action collects illness symptions 12

Contact Solutions Venture Investment award 8 Syntellect CIM 8.0 13

Convergys expanding call centers 8 Syntellect/Oracle alliance 13

DiRAD at Minnesota 9 CT Connect incentives extended 14

PCI guidance for CC processing 9 Venstar ComfortCall 14

Eckoh Utility payment service 9 Verint adds to Audiolog for contact centers 14

Alcatel-Lucent's Genesys/Enterprise Integrated

10 NG9-1-1 support from Verint 14

RoomPlace Selects Interactive Intelligence 10 Leading Software Selects Voxeo's Cloud 15

KnoahSoft using Aurix 10 VoxGen obtains UK grant 15

Nexidia at BC/BS Tennessee 11 VXi TalkPro Speech Recognition Headsets 15

Telephone Messaging - Applications/Products/ Partnerships/Financial Adeptra recognized by FST magazine 16 Movius expands messaging services 17

Aastra BluStar Video Conferencing phone 16 Nuance Powers Vivo Voicemail-to-Text Service:

18

AVST at ACUTA 16 West launches Campaign Central 18

Jott shuts down 17

Telephone Tools - Applications/Products/ Partnerships/Financial

IVONA for Developers is doing well 18 VBVoice provides MS MRCP Connector 19

I6NET/IVONA VoiceXML browser or Asterisk 19 Voxeo Adds Mobile Web Access to Cloud-based VoiceObjects On-Demand

19 Loquendo/NuEcho partner 19

Page 2: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 2

Embedded Mobile – Applications/Products/Partnerships/Financial

ZoomReader App Released for iPhone 20 Parrot Asteroid Brings Internet To The Car 24

Play Antakshari on Mobile 21 Quickoffice for Android Tablet Devices 24

Taskmind iPhone app from Catalysts 21 SoundHound adds recommended songs 25

Cyberon Voice Speed Dial for Android 21 SpeechCycle Launches SmartCare Mobile 25

Dial2Do Powering SuperTooth Assistant 21 TeleNav hybrid navigation 26

Epocrates adds ASR to mobile EMR 22 VoiceBox in TomTom GO and VIA Products 26

Google Voice Search available in Malaysia 22 Vlingo Launches Its Virtual Assistant in UK 26

Sprint Offers Google Voice Apps 22 Voice Assist Partners with Jabra 27

NHN and Hyundai telematics partnership 23 VoiZapp\ Friends Aloud iPhone app 27

iSpeech Releases DriveSafe.ly on iPhone 23 Weather Radio App 27

Warid Congo Voice SMS on Kirusa platform 23 Yahoo search for iPhone and iPod touch 27

MobiFlex for Google Apps 23 Yap now offering Spanish voice-to-text services

27

Carolina West Selects Movius Voice-to-Text 23 Yap Free VM-to-Text for AT&T, T-Mobile and Verizon

28 Nuance Dragon apps in China iTunes store 24

Nuance Powers Movistar’s V-to-Txt Service 24 ZoomSafer additional $1.1 Million financing 28

Embedded – Applications/Products/Partnerships/Financial Bark recognition 28 Loquendo in gadget TV show 29

Garmin navigation for aircraft 29 Microsoft Kinect achieves record sales rate 29

Loquendo in Spanish Transportation system 29 NICE Recording Technology at DFW Airport 29

Desktop – Applications/Products/Partnerships/Financial Babylon 9 Will Speak Acapela 30 LinguaSys translation applications for IBM 32

Acapela in E-Doceo Online E-Learning Service

30 Accentus acquires Mrecord 32

ClearOne Unveils New Chat Speakerphones 30 M*Modal Speech Understanding in WebChartMD

32

NoteSwift debuts 30 Nuance Healthcare Development Platform 33

Nattiq Dolphin Pen Adds Acapela's Arabic TTS

31 Nuance’s PowerScribe DIACAP Certification 33

Hamilton Relay Provides Captioned Telephone Services in Virginia

31 Philips Voice Tracer 600 34

Scribe in Ambulatory Transcription Market 34

Kopin Golden-i 3.5 Computing Headsets 31 SpeechGear expands translation Capabilities 34

Google adds speech recognition to Chrome 31 Highlands and Islands go digital with letters 35

IVONA TTS in The Talking TV Guide 32

Industrial Voice Technology – Applications/Products/Partnerships/Financial

Aldata Voice Directed Warehousing 35 Psion debuts new handheld 36

Datalogic Mobile rugged mobile computer 35 RedPrairie to include Vocollect VoiceArtisan 36

Roche, Merrill Corp use Datria Voice Solutions

35 topVOX's BlueMaster Headset 36

Datria offers device-agnostic voice solution 35 Intermec, Vocollect plan integrated offerings 37

Dematic at Crown Equipment Alliance 35 Portable voice from Voxware 37

RedPrairie/Lucas Systems partner 36 Wavelink Demonstrates Speakeasy adds 37

LXE rolls out vehicle-mount computer 36 Merck using Zetes 38

Speech Technology – Applications/Products/Partnerships/Financial

APH News, April 2011 38 March, 2011 Speech Technology Patents Awarded by US PTO

39 AT&T at CSUN 38

Nuance shares rise on Apple iPhone report 38

Speech Industry People – 38

Page 3: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 3

Financials – 43 Table F1 - Q4 Quarterly results 43 Table F2 – Stock Price – 4/01/2011 43

Career Opportunities in Speech Technology – 44

SpeechTek Jobs 44 Contact Solutions 47

Calendar – 48

Nuance anti-trust violations continue: Nuance got into the IVR hosting business when they acquired BeVocal in 2007. Prior to the acquisition, Nuance partnered with IVR hosting service providers by providing base technology licenses and professional services. After the acquisition, Nuance continued as a supplier but also became a competitor to the hosting service providers. As a supplier, Nuance continued to raise the prices of its licenses and services. As a competitor, Nuance has behaved in a fashion that clearly violates anti-trust laws. Nuance has gone directly to the customers of their IVR hosting partners with an offer of per minute pricing that is a small fraction of what the partner was charging. Since Nuance owns the technology, they do not have the license/support expense that their competitors have for the basic speech technology. This violates anti-trust laws from a number of perspectives:

1. It is predatory pricing and designed to drive their competitors out of business. 2. It is discriminatory pricing, since the pricing being offered to entice an enterprise to switch to

Nuance is not offered to all of Nuance‘s clients. 3. Nuance‘s behavior is ethically bad and demonstrates a fundamentally lack of integrity. 4. With no competition, Nuance is free to consolidate control of the industry and charge whatever

prices they wish. 5. No motivation exists for further technological research, since there are no competitors left to

gain an advantage over. Despite the claims that Nuance makes about being the leader in speech-enabled IVR, the reality is that Nuance has not been very good at implementing quality self-service deployments. Just call Nuance‘s own customer support line (800-654-1187) for validation of how bad it is. Nuance implementations have historically been among the worst that exist. Telephony and database connectivity are major skill sets that are essential for having a quality IVR and these have always been in short supply at Nuance. This means that the enterprises that are the targets of the Nuance anti-trust violations will end up losing big. The system reliability will deteriorate. Nuance support has historically been expensive and poor and Nuance continues to cut back in this area. In contrast, the support and reliability invariably provided by the incumbent hosted service provider is top-notch. Finally – the pricing that made Nuance attractive in the first place will disappear as soon as Nuance has the enterprise in a locked-in position where they can raise their prices. The above is a best-case scenario for the enterprise. It ignores the possibility that Nuance will get nailed for anti-trust violations. It ignores the reality that Nuance has been unprofitable overall and highly unprofitable in the call center area. The Nuance call center business is financially unstable and its long-term viability is questionable. Revenue for the Nuance enterprise business has been declining during the last year. The monopolistic behavior of Nuance continues to hurt the industry. The self-proclaimed leader provides virtually no leadership and any of the significant issues that exist. Technology advancements provided by Nuance are virtually non-existant.

TTS in the call center:

Text-to-Speech (TTS) should be used for all of the IVR dialog. It is much more flexible and cost-

effective than recorded speech. It is much easier to create and maintain. Despite all of the benefits of

TTS, it is rarely utilized for anything other than variable data. This is just another example of the

backwardness of the speech IVR area. The basic industry view is that callers do not like TTS. Not a

single bit of data exists to support this view. In fact – just the opposite. When asked, callers make it

clear that they think the voice is just fine.

The real reason that recorded speech is extensively used in the call center is that it is a major revenue

source for the suppliers. Could you imagine someone like Nuance opting for lower professional

Page 4: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 4

services fees? The entire process, from obtaining the talent, matching the talent to the application,

coaching the talent, maintenance, updates, etc. is an expensive and time consuming one.

Using the voice to achieve a unique brand is another argument that the speech industry uses. TTS

actually gives you a broader choice of voices. In reality, considering the state of most speech self-

service implementations, a high level of concern about the voice seems like an inappropriate

consideration. The voice needs to be clear, understandable, pleasant sounding and appropriate for the

application. The ―unique brand‖ is a laugh. I hear the same voice when I call Amtrak, my health

insurance provider, my bank, etc.

Moving from recorded speech to TTS would improve the quality of telephone self-service and

significantly reduce the cost of implementation.

Speech in the Browser:

There is an effort to add speech to web browsers, which would enable web programmers to include ASR and TTS in their websites. For example, users could fill out online forms with speech instead of typing. While I don't think this effort will get people to talk to their desktop computers, I do think it has the potential to generate more applications for accessibility and mobile websites. 1) Accessibility - there won't be any excuse for a web site to not be fully accessible in the future. Every web page with input fields will have a speech option, although I don't know how the ability to correct errors will be made accessible, which is a necessity given the inevitable errors in recognition. 2) Mobile computing - HTML5 will make it easy for web sites to have speech command and control built in to them, and hands-busy uses make sense. The World Wide Web Consortium (W3C, http://www.w3.org/) is the group that defines Web standards, such as Hypertext Markup Language (HTML). The W3C is currently developing the next version of HTML, called HTML5. One of the groups within W3C is the "HTML Speech Incubator Group" at the W3C (http://www.w3.org/2005/Incubator/htmlspeech/). The mission of the HTML Speech Incubator Group, is to determine the feasibility of integrating speech technology (ASR and TTS) to provide a high-quality, browser-independent speech/multimodal experience. Initiating members of the Speech Incubator Group include Voxeo, Microsoft, Openstream, Google, AT&T, and Mozilla. Google has included a draft version of the ASR capability in the latest Beta of their Chrome browser (http://googlechromereleases.blogspot.com/2011/03/chrome-beta-release.html). If you're running the Chrome Beta, you can try out the ASR functionality here: http://slides.html5rocks.com/#speech-input. I've attached a screen shot of what it looked like when I clicked on the microphone icon and said "Hey Walt, this is Jon." As you can see, ASR will need correction unless I decide to change the spelling of my name...

Jon Staebell, [email protected] staebell.tel (608) 729-5925

Page 5: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 5

Voice Technology in the Supply Chain (VSC) market report: Tern Systems, Inc. announces the immediate availability of the 2011 Voice Technology in the Supply Chain (VSC) market report. A CAGR of 22.6% is projected for the VSC market for the period 2010-2015. This is being driven by the recognition that ROI payback periods are extremely short, based on improvements in productivity and accuracy that are achieved with VSC. The Voice Technology in the Supply Chain market is at a very early stage. The penetration of voice in distribution centers that would readily benefit from voice technology is approximately 8%. Voice Technology in the Supply Chain: Markets, Products & Suppliers 2011 analyzes the market from 2008 through 2015. The report breaks down the market segments to provide concise, detailed market data from a number of key perspectives: o Voice picking compared to other picking approaches (workstations shipped and installed base), o end-user revenue, o work-station shipments and installed base, o site shipments and installed base, o VSC by warehouse size, o Workstation shipments/installed base by type (dedicated/proprietary, multi-modal, IP telephone) o vertical industry (3PL, Consumer goods, Food/Beverage Distribution, Grocery Chains, Restaurants/ Convenience stores, Healthcare/ pharmaceutical Distribution) o application (voice picking vs other voice applications) o geographic penetration, o distribution channel, o type (expansion, greenfield or replacement) A breakdown of the value chain is presented for the period 2008 through 2015. The distribution warehouse market is categorized by size of warehouse as well as the average number of workers per warehouse. A compilation of all of the active Voice Technology in the Supply Chain patents that have been issued by the USPTO is included. Voice Technology in the Supply Chain: Markets, Products & Suppliers 2011 is the most comprehensive report that is commercially available for the Voice Technology in the Supply Chain market segment. We offer a money-back guarantee that this is so. Voice Technology in the Supply Chain: Markets, Products & Suppliers 2011 is available from Tern Systems. Contact Walt Tetschner at [email protected] or 978-266-1966. Figure 1.5b Total VSC Work Stations Installed Base as % of Total Available Market 2008-2015

Warehouse Workers using Voice

0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

16.00

2008 2009 2010 2011 2012 2013 2014 2015

%

Page 6: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 6

Some real-world views on speech

OnStar‘s new Facebook app:

http://gadgetwise.blogs.nytimes.com/2011/03/25/a-23000-facebook-app-that-comes-with-a-car/?src=busln ASR replacing the t in boots with a b was funny. http://www.sync-blog.com/sync/2011/03/voice-recognition-is-it-reliable-for-everyday-use.html

Speech Technology Patents issued last month: During February, 133 speech industry patents were issued by the US PTO. AT&T and Microsoft had 11 each, IBM had 9, and Nuance and Sony had 7 each. Health Hero Network and Avaya had 3 each. Intel, Jumptap, Samsung, Securus Technologies and Verizon had 2 each.

Nuance accent patent: Nuance was awarded a patent titled: System and method for automatic speech recognition 7,912,721 last month that covers an interesting new way to handle accents. Regional accents occur frequently in most languages. In most cases, this is caused by certain sounds being left out by speakers from a particular region. For example, natives of New England do not pronounce the letter r sound. To accommodate them, alternate pronunciations have to be included in the vocabulary and/or they need to be represented in the speaker database. This approach tends to have the effect of degrading the overall recognition performance. The method that is described in this patent automatically recognizes a regional or non-native accent and then makes a substitution that modifies the utterance to be compliant with the standard method of pronouncing an utterance. It is then sent to the recognizer. The techniques described in this patent seem to have the potential for automatically accommodating non-native speakers, without degrading the overall recognition process.

Get2Human.com Get2Human Great Customer Service Club

adds to Utilities segment: The Get2Human Great Customer Service Club consists of enterprises that have been selected as having superior customer service. This month we've added Central Maine Power to the Utilities that have been selected for membership in the Get2Human Great Customer Service Club.

Thought for the month:

Q. You're trapped in a room with a tiger, a rattlesnake and a lawyer. You have a gun with two bullets. What should you do? A. Shoot the lawyer. Twice.

Page 7: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 7

Telephone IVR – Applications/Products/Services/Partnerships/Financials

Atos Origin Awards Voice Biometric Contract to Leading Software: Loquendo, Atos Origin, and Leading Software announce that Leading Software - leveraging the latest voice biometric technologies from Loquendo – has been chosen to provide a secure 24 hour, 365 days a year service to Atos Origin‘s key financial service and public sector clients. Atos Origin, the international IT services provider currently provides help desk support to over 1200 Global blue chip companies with over 12M call annually and over 1M Password Resets per annum. Recent research by Forrester shows the average cost of a password reset is £15 with Atos Origin reporting an average of three to four password resets per employee every year. The Leading Software Identity Management Self-Service PIN/Password Reset solution identifies the caller and subsequently allows them to reset their password or unlock their account without the need for help desk interaction. Loquendo Speaker Verification makes use of the voiceprint as a biometric measure to verify an individual‘s identity. Accurate, versatile, simple to use and non-intrusive, it results in high user acceptance and requires no specialized equipment - just a phone. The ideal solution for resolving security issues around remote telephone access to many types of services, Loquendo Speaker Verification can be used in a wide variety of application contexts, such as phone banking and trading, password resetting, accessing customer care services, credit card activation, transactions and payments. Conducting rigorous analysis of Atos Origin‘s UK operations, Leading Software installed and integrated employee enrolment of a voice based biometric self-service password system into existing systems with no personal data associated with the user‘s voiceprint. With a significant and fast return on investment for minimal outlay, Leading Software‘s self-service voice based biometric solution substantially reduces Help Desk overheads and delivers considerable productivity gains while offering staff cost savings to Atos Origin‘s clients. The Self-Service/PIN Password Reset is one of a number of voice biometric based identity management solutions and services from Leading Software Identity Management including User Authentication via Call Centre or IVR and Mobile Phone Banking

Aurix announces market availability of its cloud-based speech analytics: Cloud Based Analyst (CBA) is an innovative software solution from speech technology experts, Aurix that has put speech analytics ―in the cloud‖, allowing the scalability, flexibility and ease of use needed to gain real business intelligence from call recordings. The product eases the demand on developers and analysts, as it allows managers themselves to analyse calls and create customised reports without any specialised knowledge. The desktop interface for audio search and view has been designed to be as simple as possible, and the upload of search results to the analytics engine is automatic. An easy-to-use report wizard allows individual users to customise and create new metrics and reports, without needing to call a developer/analyst. Plus, Out of the box reports mean fast start up and reduce deployment time and cost. There is even a mash-up facility, allowing users to upload data from customer surveys, to give comparisons between what customers think and what agents are saying. The development of the software in the cloud means CBA is infinitely more flexible and scalable than traditional systems and can be implemented and maintained at a fraction of the cost. CBA uses Aurix‘s existing gopher-it desktop tool, which is linked to a cloud-based analytics tool, supported by a data warehouse – Aurix partners with Cloud Analytics provider GoodData, to deliver the service.

Avaya partners develop UC apps for SMEs: Avaya announced an array of application development partners driving new and innovative capabilities for Avaya IP Office, the company's flagship communications solution for small and medium-sized enterprises (SME). These

partners are part of Avaya's DevConnect program, an initiative to promote the creation of a new generation of solutions complementary to unified communications (UC) by integrating products, services and technologies with Avaya solutions.

Page 8: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 8

Hospitality Applications Enhance Guest Service for Hotels The DuVoice DV2000 is a messaging system and Property Management System (PMS) interface for Avaya IP Office that is suited for hotels. DV2000 capabilities include everything from voicemail and wake-up call management to enabling employees to handle mini-bar inventories via phone. DuVoice also provides 911 features for Avaya IP Office that alert the front desk when emergency calls are made and pinpoint the location of the room requiring help. Contact Center Solutions for Streamlined Customer Service KnoahSoft complements and enhances Avaya IP Office's customer service capabilities using the KnoahSoft Harmony suite, which offers secure call recording, silent monitoring, screen capture of agent sessions, as well as extensive quality management and reporting capabilities based on search parameters. By bringing these capabilities to Avaya IP Office and providing legal compliance support for contact center recording, KnoahSoft extends state-of-the-art customer service to the SME environment. Automated Applications Simplify Processes Computer Instruments offers e-IVR, which contains the components that allow companies, schools, churches and others to reach out to constituents automatically to remind them of appointments, inform them of school cancellations, and alert them of emergencies. These messages may be sent via phone, email or text by seamlessly integrating with Avaya IP Office. Inbound functions are just as useful, allowing callers to find out about homework that is due, renew a prescription, or pay their utility bill--all via the simplicity of speech recognition or phone prompts.

Trident Marketing, Dimension Data Partner to offer New Communications Platform: Dimension Data announced that Trident Marketing

enlisted Dimension Data‘s customer interactive solutions group to design, build and support its customer service platform. Trident enlisted Dimension Data to deploy Cisco Customer Voice Portal (CVP) – a carrier-class Voice over IP (VoIP) interactive voice response solution. Dimension Data implemented the Cisco CVP solution with Trident‘s SIP trunking architecture.

Contact Solutions receives Venture Investment award: The Contact Solutions model of continuously improving the solutions they provide to their customers – increasing their IVR usage while improving the customer experience – is one that has helped us create strong business relationships with many companies. It is Contact Solution‘s unique commitment to continuous improvement and customer experience that

attracted North Bridge Group Equity to invest in Contact Solutions last year. That investment was recently recognized by the Washington Business Journal in its Venture Capital Awards. Contact Solutions received recognition as a runner-up in the Deal of the Year for Later stage companies along with North Bridge.

Convergys expanding call centers:

Convergys Corporation, the largest private employer in the Philippines, announced that it would expand

again, with a fourth state-of-the-art facility in Cebu City. The announcement comes just weeks after Convergys revealed it has a new contact center currently under construction in the country‘s summer capital, Baguio City. Located in the Asiatown IT Park, the new facility being announced today is at the TGU Tower. Once the Convergys site is operational, the company will operate 15 facilities throughout the country - eight in Metro Manila, four in Cebu City, and one each in Bacolod City, Santa Rosa, Laguna, and Baguio City. Convergys also announced plans to hire agents in: Cincinnati; Salt Lake City, Waco, Orlando, Ft Lauderdale, Logan, and Chattanooga.

Page 9: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 9

DiRAD Awarded Minneapolis Cloud IVR Contract: DiRAD Technologies, Inc., a 27 year-old Voice Solution Provider, was awarded a contract to design, implement and host a web-based Interactive Voice Response (IVR) system for the City of Minneapolis, Minnesota. Leveraging speech recognition for both English and Spanish-speaking customers, the system allows callers to inquire about their water service billing, get balances, payment history, and avoid shutoff, 24 hours a day. The system also provides Minneapolis staff with an outbound notification application to alert customers of billing issues and pending water shut-offs. Staff can access the system entirely from the web, with tools administration, reporting and real-time call traffic statistics. Minneapolis Project Manager Jeff Kendall said, "I have been a project manager for many years and in that role I have worked with many Vendors. DiRAD has to be one of the best I have encountered. The City of Minneapolis has many processes in place to monitor/manage projects during each step through the project life cycle. The DiRAD team stepped up and did not only what was asked of them, but also provided suggestions on how we can improve." The system integrates with the City's Enquesta Utility Billing platform, provided by Systems & Software of Williston, Vermont. PCI Council Releases Supplemental Guidance for Protecting Telephone-Based Payment Card Data: If you handle credit card data, you fall within the scope of the Payment Card Industry’s Data Security Standard (PCI DSS). That‘s the central message of a new supplemental guidance document the PCI Security Standards Council released today. The Protecting Telephone-Based Payment Card Data Information Supplement provides actionable recommendations to merchants and service providers for securely processing payment card data over the telephone. It also offers advice for protecting cardholder data in recorded transactions. The council developed the information supplement to assist merchants and service providers with meeting PCI DSS requirements to secure payment data captured within voice recordings. Its main message is simple: If you don‘t need it, don‘t store it. And if you do need it, store it for the minimum amount of time possible and make it unreadable with some sort of truncation and encryption. The PCI call center guidance includes the following: Explanation of how PCI DSS applies to cardholder data stored in call recording systems, with detailed tables that map types of data to specific PCI DSS requirements. Recommendations for merchants when assessing risk and applicable controls of call center operations, with a quick reference flow chart that provides a step-by-step process for determining necessary controls to meet PCI DSS requirements for voice recordings. Specific guidance addressing storage of sensitive authentication data, including suggested methods for rendering data unavailable under query to meet PCI DSS requirement 3.2. Guidance on some of the key considerations faced by call centers when implementing PCI DSS requirements, including specific recommendations and best practices.

Eckoh supports mobile phone ‘top-up’ utility payment service: Eckoh announced the expansion of its three-year contract with Utilita Group to include the provision of an additional payment capability for mobile phones. Utilita supplies gas and electricity. Utilita customers that have pre-pay meters can purchase electricity or gas using the EckohPAY SMS service. Customers register their mobile number online and when they want to ‗top-up,‘ they send an SMS message with their electric or gas card number and the amount they wish to pay to a dedicated shortcode. Once the details are validated and authorized, the customer is issued a code to activate the purchased electricity or gas.

Alcatel-Lucent's Genesys and Enterprise Integrated: Alcatel-Lucent has brought together two of its solution portfolios – Genesys and the Enterprise – in a bid to expand its Dynamic Enterprise vision and increase its

market reach. the convergence of Genesys call center activities and enterprise product lines will enable them to deliver a ―powerful‖ integrated suite of solutions.

Page 10: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 10

To enterprises, Alcatel-Lucent‘s evolved Dynamic Enterprise vision offers a path for building sustained strategic differentiation by interconnecting their networks, people, and processes and harnessing the connected knowledge of their customers, employees, and partners. The company‘s Dynamic Enterprises approach brings a new method to enhance ―collaboration‖ with cross-channel ―conversation.‖ In addition, it also boosts employee and customer conversation management over time and across any media, in a more Web 2.0 approach. Specific solutions are expected to be introduced later in the year. Moreover, the new approach focuses on taking the enterprise communications platform and driving it to a more software-based architecture. Also, it aims to make it open to third party developers to create new, mashed up services and applications. As part of the Dynamic Enterprises strategy, Alcatel-Lucent has unveiled a new Smart DeskPhone, featuring an open API developer platform, multimedia communications capabilities, and Web applications. All these features together enable the creation of personalized mashups or industry-tailored applications.

The RoomPlace Selects IP Business Communications Solution from Interactive Intelligence: The RoomPlace has selected the Interactive Intelligence all-in-one IP communications software suite, Customer Interaction Center™ (CIC), for use throughout its organization. The RoomPlace is replacing communications technology from multiple vendors with CIC to connect all of its 24 locations. After evaluating solutions from Avaya, Cisco, Nortel and Shoretel, The RoomPlace selected CIC because of its all-in-one software architecture that offered the ability to mix and match a broad set of features with select

user groups. Following the initial deployment, the company plans to roll out CIC to the rest of its 900-plus employees located across 24 locations throughout Illinois and Indiana. The RoomPlace will use CIC applications for IP PBX, automatic call distribution, interactive voice response, screen-pop, unified messaging, presence management, conferencing, faxing, and real-time monitoring. It will deploy CIC add-on applications for multichannel recording, outbound dialing, and automated customer feedback surveys.

KnoahSoft Includes Aurix's Detailed Speech Analytics in Harmony: All of the Harmony Editions can be enhanced with speech analytics. Harmony Speech Analyzer, powered by Aurix, enables call center supervisors and management to perform "precision monitoring" on keywords or phrases, using the smart Phonetic based search. Harmony Speech Analyzer Feature Highlights: Identify important business trends, customer insights, revenue opportunities and competitive situations Easily create KPIs, associated words and phrases to increase customer satisfaction Report on agent scripts adherence. Automatically zero-in on problematic calls in to identify potentially serious and costly issues and/or a need for specific agent coaching. Mine calls for required or forbidden words and phrases for any marketing, quality control, compliance, or general interest reasons. Detect periods of unwarranted silence which drives up AHT Various alerts and notifications to help supervisors & managers take corrective actions.

Nexidia Speech Analytics Identify and Promote BlueCross BlueShield Tennessee's Contact Center Wellness: BCBST ‘s BlueCard division is responsible for linking independent Blue Cross companies across the country. And to serve them, the healthcare firm has about 80 dedicated contact center agents who field over 400,000

inbound calls per year. A few years ago, however, BCBST began noticing that the BlueCard division‘s contact center

team was showing some disturbing symptoms: that its call volume and average handling times (AHT) was continuing to spiral far above average and at alarming rates. While the overall organization

Page 11: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 11

experienced a nine- to 10 minute- AHT, the BlueCard team saw theirs grow to an average of 17 minutes: more than 70 percent higher.

While BlueCard could analyze AHT, the diagnostic tools it had were limited; it could only do so at an aggregate level. That meant that despite having multiple ways to improve AHT, including added training and improved agent scripts to developing online application BlueCard could not pinpoint which solution mix would address the issues.

Instead the BlueCard team mainly relied on talking and listening to the patients, i.e. contact center agents. It could not see deeply enough to find out what was causing the illness -- why customers were placing their calls. Without this ability to probe into the interactions to identify trends and understand the root causes, the BlueCard group did not know which treatments to administer that would cure or mitigate the conditions. Yet having this information could reduce its call volume, improve service metrics such as first call resolution (FCR), and boost operational efficiency.

BlueCard found a promising diagnostic methodology, Nexidia‘s QuickStart program, and introduced it into its practice in February 2009. Working with the BCBST BlueCard team, the Nexidia analyst assigned to the account quickly understood the topics customers were speaking about, and what operational triggers were causing specific customer calls. They could then probe deeper into the underlying conditions and analyze the results. When they did, they found these issues, which were then treated: The long AHT was caused by many incoming provider calls indicating they had sent BlueCard medical records that could not be found in the system. With Nexidia, BlueCard discovered that a simple process issue prevented agents from easily locating records. Missing provider cover letters were preventing records from being automatically imaged into the appropriate system for future retrieval and routing. With this knowledge, BlueCard then identified those providers who were not sending their medical records appropriately and alerted them to review the appropriate filing protocol. The Nexidia solution also informed the BlueCard team to the need for specific types of training for its contact center agents. To boost FCR and improve group claims handling: assigning them to the appropriate agents for follow up, BlueCard needed to provide clearer training to its agents on the classification of specific call types. BlueCard also identified where its website could be improved. Using Nexidia, BlueCard identified that the billing offices did not always receive important paper remittance advices that would be helpful in later conducting a claim status online. From this, BlueCard improved self-service options by enhancing its Web site to provide a link to the relevant remittance advice each time a provider updated a claims status. The results with the Nexidia hosted speech analytics were quick to achieve, tangible and measurable. AHT dropped to 12 minutes from 17 minutes -- a 42 percent improvement -- resulting in an estimated nearly $900,000/year savings. Call volume dropped 12 percent for calls related to ―medical records needed to process claims not located,‖ saving an estimated over $130,000/year. Improved self-service options by updating its website to include links to updated remittance advices also reduced costs by diverting calls. FCR was enhanced through managers employing the speech analytics technology as an agent training tool on specific types of calls. For example, BlueCard built queries around a forwarded call and work with staff that needed extra training on that scenario, thus helping to improve FCR. The staff was receptive to the opportunity for improving specific areas of their skill sets and witness efficiency gains, further enhancing FCR. In just four months BlueCard recorded a return on investment of over $1 million.

Vormetric Provides Data Security for Nexidia’s Hosted Speech Analytics Operation: Vormetric announced that Nexidia has selected Vormetric Data Security to provide encryption, access control and key management to meet regulatory compliance requirements for PCI DSS, HIPAA and

other mandates that govern its OnDemand hosted solution. The Nexidia OnDemand platform provides contact centers access to the full set of features in its Enterprise Speech Intelligence product suite (ESI), including automatic topic discovery, detailed call categorization and reporting, and agent performance management. Since inception, the Nexidia OnDemand hosted platform has grown significantly and now processes more than 150,000 hours of audio per day for large contact centers and outsourcers.

Page 12: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 12

Nexidia selected Vormetric to provide data security for this large volume of audio files and to help comply with customer security requirements. After a comprehensive evaluation of competing encryption products, Nexidia selected Vormetric Data Security to protect millions of files accessed by Nexidia ESI running on high performance servers. Nexidia chose Vormetric for its ability to centralize and simplify key management, enforce policy, impose virtually no impact on application performance, and be extremely low maintenance.

New Hampshire hangs up on its 511 system: New Hampshire‘s has shut down their 511 telephone service. The primary reason was to save money while the economic downturn continues. The Nashua newspaper incorrectly reported that it was shut down because of ASR errors. The original 511 system that was installed did, indeed, work poorly. It was based on an out-of-date system that was provided by a supplier (Frank Solutions) that no longer exists. This problems with the ASR were solved a number of years ago and the service has been running on Nuance On-Demand (NOD). The ASR error aspect made the story more sensational, but it was incorrect.

Nuance Takes Natural Language on the Fast Track: Through Call Steering Fast Start, Nuance now offers a scaled-down version of call steering and natural language for companies with smaller call volumes. Using Nuance‘s SpeakFreely technology, the solution not only costs less on

the front end but promises a quick payback on the back end through reduced contact center costs associated with complex IVR menus, misdirected calls, and agent-to-agent transfers. Nuance still offers its full-service package, but Fast Start is priced to fit into smaller budgets, easier to scope in pre-sales discovery, and faster to deploy. Instead of the 20,000 initial utterances and 50,000 additional utterances after pilot tuning that are included with Nuance‘s full-service offering, Fast Start works with 3,000 to 5,000 initial utterances and scales to about 20,000 after pilot tuning. The Fast Start package also factors in a lack of excessive business logic for call routing and no personalization or data dips, a pre-defined call flow, and fewer than 100 possible destinations within the menu. The offering not only cuts the price of a natural language call steering application in half, but it also cuts the time to deploy in half.

Nuance and Cyara Solutions Partner to Ensure Successful Deployments of Speech and Voice Biometrics Solutions: Cyara Solutions announced it has signed an agreement with Nuance Communications, to resell Nuance‘s Enterprise software products in conjunction with Cyara‘s products, to deliver better voice self service experiences for customers in Australia and New Zealand. Cyara provides an end-to-end solution for testing IVR applications that utilize speech recognition, text-to-speech(TTS) and voice biometrics. The Cyara Solution Suite tests IVRs, speech platforms and contact centers from initial deployment through subsequent system and application modifications, fine-tuning and ongoing operations in production to ensure successful deployment or upgrade, exemplary system operation and customer experience.

Smart Action collects illness symptoms: David Kirsh, professor of cognitive science at UC San Diego, working with a grant from the Centers for Disease Control and Prevention, has chosen SmartAction Virtual Agents to capture and report on student illness symptoms at an elementary school in 4S Ranch. Smart Action provides automated virtual agents powered by artificial intelligence that answer and make calls for contact centers, company officials say: 'The system also features natural language speech recognition and synthesis technology.' The pilot project was established to determine if increased information about medical absences reported to schools would lead to improved intelligence about contagious diseases, possible biological attacks or other notable trends in student health, Kirsh said: 'This type of system greatly improves information quality and enables more rapid analysis, allowing us to detect trends in symptoms and disease.'

Page 13: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 13

When parents call the school to report children absent or ill, the SmartAction Virtual Agent answers the call and asks a number of questions about symptoms. Information provided goes into a database for later reporting and analysis. The Virtual Agent identifies parents or relatives based on Caller ID, and remembers yesterday‘s symptoms for speedy reporting. The system also gives callers the option to provide additional information, which is recorded. The results are then reported to researchers either by e-mail with a summary of information captured, including recorded parts, via a real-time database update to the university system or by a Web-based report with a click-to-listen feature. 'After a poor experience trying a conventional interactive voice system, it was clear we had to find something more advanced to do the job,' Kirsch said. 'Smart Action had the technology to provide an optimal caller experience and capture the needed information.'

Syntellect to Release Version 8.0 of CIM Solution: Syntellect announced the upcoming release of Syntellect Customer Interaction Management (CIM) version 8.0 from its line of

customer interaction solutions. Syntellect CIM version 8.0 includes a number of new product enhancements, including: New auditing feature helps ensure compliance and logs a complete audit trail with reports of system changes made by agents, supervisors, and system applications Integration with Syntellect Communications Portal (SCP) for comprehensive handling and management of all voice interactions on a standards-based voice portal Agent and resource manager module updates increase usability and provide more flexible desktop integrations iVault now has partitioning of the interaction views to ensure that Agents only have visibility to the appropriate data Decision Manager enhancements enable more flexible reports New localization kits for Configuration Manager, iVault and QIM Updated support for many third-party components Additional features available in the second half of 2011 include: inbound and outbound SMS message support through native interfaces as well as third-party aggregators, client user interface overhaul using thin client technologies, Supervisor/Agent Messaging enhancements, and support for VMware ESXi.

...

Syntellect and Oracle Partner to Deliver Unified Communication Solutions: Syntellect has established a strategic CTI partnership with Oracle,

the world‘s most complete, open, and integrated business software and hardware systems provider. This engagement will focus on each organization‘s extensive strengths and further establish CT Connect as a proven, robust, and certified CTI solution for the Oracle E-Business Suite. Oracle E-Business Customer Relationship Management (CRM) solution is an integrated set of applications that provides information-driven sales, service, and marketing. Its constituents Oracle® Advanced Inbound Telephony and Advanced Outbound Telephony allow integration of telephony infrastructure with the contact centers. Syntellect CT Connect for Oracle E-Business Suite provides integration between Oracle Advanced Inbound Telephony and Oracle Advanced Inbound Telephony and a wide variety of telephone switches found in the market. As part of this partnership, Oracle and Syntellect will focus on building and certifying the adapter between the latest releases of CT Connect and Oracle E-Business Suite. Oracle already provides a pre-built and tested adapter for integrating the Oracle E-Business Suite R12 and Syntellect‘s CT Connect 6.1 product line which enables the agents to quickly understand the context of the call through screen pops with relevant information about the current customer contact. Click-to-dial functionality allows agents to dial contacts out of the business applications with minimal clicks. All the customer interactions over the phone are also automatically logged.

Page 14: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 14

Special CT Connect Incentives Extended to April 30, 2011! Syntellect CT Connect is the industry-leading solution that has CTI

enabled more than 1 million agent seats worldwide. Now, with promotional server-based pricing, you can add CTI functionality to your computer-telephony

infrastructure, including free adapters for popular CRM solutions Salesforce.com and Siebel, for as low as $50 per seat. Save even more when you purchase an unlimited license for CT Connect. In addition to server-based pricing, special Salesforce.com and Siebel subscription pricing is available starting at just $12 and $17 per month respectively. This subscription price includes maintenance and requires a 10 seat and 12 month minimum. These special incentives have been extended to April 30, 2011.

Venstar ComfortCall: Venstar announced the availability of ComfortCall, an accessory that allows users to remotely adjust their Venstar Slimline residential thermostats from any telephone. Using voice recognition and voice synthesis, ComfortCall literally talks to the user over the telephone and takes voice commands. ComfortCall has a small, attractive attachment that plugs into the bottom of any Venstar Slimline thermostat. It allows users to call and literally talk to their thermostats from any telephone. ComfortCall will tell the caller what the inside and outside temperatures are, the status of the thermostat (if the heating or cooling is currently on or off), and will allow the user to adjust the thermostat to its pre-set comfort settings or energy saving settings or turn the system off. This is all done via voice or keypad entry from any telephone. Expert mode makes making changes very fast and easy for the user. In literally seconds, he or she can call and tell ComfortCall what to do. In addition to the small attachment on the thermostat, ComfortCall has a small wireless base station, about the size of a deck of cards, which connects directly to the telephone line or answering machine. It is fully compatible with answering machines, utilizing a command that sends the call to the ComfortCall thermostat attachment and hangs up the answering machine.

Verint Advances SMB Contact Center Recording and Quality Solution, Adding New Functionality to Meet Changing Market Demands:

Verint Systems announced the availability of the latest version of Audiolog for Contact Centers—its advanced, digital multimedia recording, evaluation, archiving and quality assurance solution. Expressly designed for today‘s small and mid-sized (SMB) contact centers, the software supports and facilitates multichannel customer interaction service delivery and introduces new security and risk management capabilities, enabling businesses to excel in serving customers across communications channels, while adhering to compliance requirements and evolving industry standards. Traditional audio, call-based service—while still a popular channel, particularly in certain demographics—is starting to give way to multichannel sales and support, driven largely by consumer demand to do business with organizations at the time, place and channel of their choosing. To equip today‘s SMB contact centers to respond to the rise in multimedia communications channels and devices, Verint Witness Actionable Solutions has added a host of enhancements to its Audiolog for Contact Centers software—which became generally available in December 2010. Contact center managers can use Audiolog‘s new multimedia replay interface to select and playback—in a timeline—multiple communications from a customer or group of customers across media types, such as calls in addition to emails and chat sessions. These multiple communications, as related to a specific issue of concern, can then be saved as a single case, for example, to facilitate more effective investigation and service/complaint follow up.

Verint Supports NG9-1-1 with Enhancements to Multimedia Public Safety Solution: Verint Systems announced a series of enhancements to its Impact 360 for

Public Safety Powered by Audiolog software, the company‘s workforce optimization (WFO) solution specifically designed to enhance accuracy and performance in today‘s Next-Generation 9-1-1 (NG9-1-1) emergency communications and public safety command and control environments. With these enhancements, the company is introducing robust incident reconstruction capabilities and support for a

Page 15: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 15

host of new requirements associated with NG9-1-1 and the rise in multimedia communications channels and devices used to report emergency situations and incidents. Among the latest enhancements to Verint‘s public safety solution, which became generally available in December 2010, are robust, new multimedia recording, storage, retrieval and playback capabilities. Now in addition to traditional circuit-switched voice trunks for audio calls, public safety answering points (PSAPs) and emergency communications centers can benefit from the ability to capture, analyze and act based on input from a wide variety of multichannel data and information sources—including traditional, wireless and radio, photos, text/SMS, videos, maps, telematics, and such related data as telephone numbers and locations. The solution‘s support for IP communications and multimedia capture—the hallmark of NG9-1-1—along with other advancements, are paving the way for improved response, stronger liability management, faster and more informed decision making, and better ultimate citizen service. Coupled with the technology‘s proven recording and quality assurance capabilities, agencies can evaluate call taker and dispatcher performance and strengthen coaching and training, particularly as new systems, processes and requirements from multichannel operations and NG9-1-1 continue to take form.

Leading Software Selects Voxeo's Cloud For Mobile Voice Biometrics Solutions: Voxeoand Leading Software announced a

partnership where Leading Software is developing voice biometrics solutions for authentication in the Voxeo cloud. These solutions include services such as password reset, mobile phone transaction authentication and more. Leading Software selected Voxeo because of its ease to use and the ability to quickly offer customers a hosted platform with a minimum fuss. Leading Software is now able to implement a hosted solution very quickly and to react fast on changing customer demands and requirements."

VoxGen Receives U.K. Government Grant to Develop Voice Biometrics: VoxGen, a provider of self-service automation solutions, has been awarded a grant of £300,000 ($481,000) from the U.K. Technology Strategy Board to research, develop, and enhance voice biometrics to support consumer identification and verification technologies. VoxGen will be working with its consortium partners, Mydex Data Services and Latitude Partners, to enhance the traditional approach to voice biometrics by adding a second factor to the verification process; personal knowledge. VoxGen will develop a statistically rigorous probabilistic reasoning system to combine these two incongruent data points. To ensure that the technological advancements from the project build both competitive advantage in the future and are exploited by business and organizations, all developments are rigorously tested in real-world scenarios with real-life users, throughout the creation process. This research determines which types of enhancements work best for government/community service users and for online retail consumers. The grant is part of a joint investment of £8.5 million by the Technology Strategy Board and the Engineering and Physical Sciences Research Council in 30 innovative research and development projects that will help accelerate the introduction of more secure and trustworthy information systems. The investment will also help to create competitive advantages for businesses through building a strong capability base in the United Kingdom.

VXi Corp Launches VXi TalkPro Speech Recognition Headsets: VXi has launched its new next generation VXi TalkPro speech recognition headsets for contact center, office and Unified Communications (UC) environments. The VXi TalkPro line includes three categories, consisting of 14 new models to adhere to the needs of any environment where employees can benefit from using a headset. All of the 14 new models include a sleek, one-piece headset with a lightweight and durable design that is comfortable enough to wear all day long. According to VXi, a new and improved microphone provides enhanced speech

recognition accuracy, ensuring that software applications receive precisely what is said, free of extraneous, error-causing noise or background interruptions. A precision gooseneck microphone boom keeps the microphone in position, maintaining consistent, proper microphone placement.

Page 16: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 16

The VXi TalkPro USB headsets feature a USB connector that plugs directly into a computer USB port. The VXi TalkPro USB headsets are available in three wearing styles: monaural (VXi TalkPro USB1), binaural stereo (VXi TalkPro USB2) and convertible (VXi TalkPro USB3) including over-the-head, over-the-ear and behind-the-neck. VXi TalkPro UC headsets incorporate all the functionality of the TalkPro USB headsets as well as advanced features including DSP (digital signal processing) and echo cancellation. The VXi TalkPro UC headsets are also available in monaural (VXi TalkPro UC1), binaural stereo (VXi TalkPro UC2) and convertible (VXi TalkPro UC3) wearing styles. VXi TalkPro headsets range from $79-$159 and are all backed by a two year warranty.

Telephone – Messaging Applications/Products/Services/Partnerships/Financials

Adeptra’s engagement with Nationwide recognised by FST Magazine: Adeptra announced that its work with Nationwide Building Society has been shortlisted in the FST awards. The highly coveted awards recognise excellence and innovation in information technology within the UK and

EMEA financial services sectors. A key area of the shortlisted ―Focus on Fraud‖ initiative involved broadening Nationwide‘s existing relationship with Adeptra, by introducing an automated interactive communication solution. This solution makes a call to the consumer when a suspected fraudulent transaction has been attempted on their card. The message gives details of the transaction in question, as well as other recent transactions and asks consumers to confirm, using the telephone keypad, whether the transactions are genuine.

Aastra BluStar 8000i video conferencing phone: Aastra Technologies Limited has introduced Aastra BluStar 8000i, a desktop telephone with a 13-inch touch screen and camera, designed specifically to facilitate video conferencing. It combines HD video conferencing quality with unified communications and business applications. The video quality is matched by HD sound quality, delivered via four voice-tracking microphones and three speakers, allowing the device to serve multiple attendees in one room. The device supports sharing content from a desktop PC, and a Bluetooth interface facilitates the transfer of data from mobile phones and Bluetooth-enabled devices. In addition to the collaboration features, Aastra BluStar 8000i provides a range of built-in business productivity applications, which can run at the same time as the live video. Half of the display can be allocated to user-defined applications. The product integrates Aastra‘s unified communications applications as well as applications from third parties. Speech recognition is supported for commands, with country-specific languages. Fingerprint reading can be used to validate users.

AVST Demonstrates Campus Mobility at ACUTA: Applied Voice & Speech Technologies, Inc. (AVST) showcased an all-new, theater-style booth, addressing key IT initiatives for colleges and universities today, including mobility, centralization, virtualization,

and deploying private cloud to public cloud UC solutions. AVST held ongoing, live presentations on how to consolidate and centralize an institution‘s voice infrastructure, virtualize communications and deploy private cloud to public cloud UC solutions. AVST also demonstrated how to drive campus mobility through features such as find-me/follow-me, unified messaging, single number access, notification and fixed-mobile convergence to keep faculty informed while on the move. AVST‘s CallXpress provides what universities and colleges require to keep up-to-date with the ever-changing world of the communications market by delivering best-of-breed UC applications, such as, unified messaging, call processing, voicemail, speech and personal assistant - all on a highly survivable and interoperable platform.

Page 17: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 17

Jott service ending on May 3rd, 2011 Following is the letter announcing the shutdown of the Jott service: By now you probably know that we will no longer be offering the Jott service. We have included below the text of an email we sent to our active customers today.

We founded Jott in April of 2006. In the five years that have followed, we‘ve been honored to serve you and work with outstanding people. We made great products. As we said in the letter below, this is really just the beginning, and we are very much looking forward to having continued, positive impact on people, but just on a far larger stage. Many thanks. We are writing to give you advance notice that as of May 3rd, 2011, the Jott.com service will no longer be available. We know how much many of you have come to rely on the Jott service, so please know that this decision was not made lightly. Please be assured that we‘ll work to help you through this transition. We truly appreciate your business and enthusiastic support over the years, and apologize for any inconvenience this may cause. In 2006 Jott emerged as a leader in mobile voice-to-text applications. Now, nearly two years after Jott‘s acquisition and a successful integration into Nuance, we will officially end the Jott service. This may seem counter-intuitive – success leading to a shutdown. But while it is an ending of sorts, the reality is that the technology, service, talent and imagination of Jott will continue on as part of a far broader set of services. The Jott team, vision and technologies are an integral part of a global business which includes partners such as AT&T, Rogers, Bell Canada, Vodafone, Cisco, Vonage, and many others. The dramatic expansion of Nuance‘s voice-to-text services is clear evidence that consumers will respond to innovations that carriers offer, and with that in mind, we will focus our voice-to-text service investments on carrier and enterprise distribution, and no longer on Jott.com. What to expect After May 3rd, 2011 you‘ll no longer be able to access any of the Jott services – Jott Assistant, Jott Voicemail, or Jott for Salesforce. Until then, Jott will continue to run as you‘ve come to expect. You will not be charged for this last month – we‘re making it free to all of our current subscribing customers, starting today, April 4th, 2011. Those of you who have paid in advance for annual or pay-as-you-go plans should expect to see a refund by June 2nd, 2011 for any unused portion of your plan as of April 4th, 2011. How to download your information that’s on Jott.com After May 3rd, 2011 you will no longer be able to access your jotts, voicemails, or contact information. Until May 3rd, you can download any jotts or contact information that you want to keep. We will send you a reminder, but recommend that you export the data that you want to keep as soon as possible via the following link:

Movius expands messaging services: Movius Interactive Corporation demonstrated a range of mobile enterprise and mobile consumer solutions designed to increase revenue and customer loyalty for mobile operators They include: Taking voice-to-text to a new, fully automated level, which includes the ability to also hear the message. Now the most comprehensive service available in the market, the Movius solution optimizes the voice messaging experience. Introducing multiple line capability on a single device. This breakthrough gives users the chance to separate their business and personal lives but still use just one device. Known as Side-Line Service, the service also provides SMS capability on the second line. Delivering simple to use audio and videoconferencing in the LTE/4G environment. This will help carriers maximize on the use of tablet PCs and smart phones and achieve additional revenue in return for their network investments. Movius Voice-to-Text comes to the market as the latest means to advance and improve the voice messaging experience. The fully automated solution can be activated via SMS therefore no manual individual provisioning is needed by the operator. Additionally, recipients have the option of listening to the message after viewing the text version by clicking a link or opening a wav file attachment.

Page 18: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 18

Nuance Powers Vivo Brazil Torpedo Recado Voicemail-to-Text Service: Nuance Communications' Voicemail-to-Text platform is now powering the new Torpedo Recado voicemail-to-text service from Vivo, a subsidiary of

Telefonica and Brazil‘s largest telecommunications carrier. Now Vivo subscribers have the ability to remain connected at anytime, anywhere, with voicemail messages converted to text and sent via SMS text message to their mobile phones. Vivo‘s Torpedo Recado voicemail-to-text service based on Nuance‘s voicemail-to-text platform gives Vivo‘s subscriber base even more mobility and flexibility, with transcribed messages delivered via SMS text message in just minutes. From there, subscribers can either delete, store or forward messages as necessary, and all transcribed messages are still available as audio voicemails by simply dialing *555.

West Launches Campaign Central for Targeted Notifications Campaigns: West announced the launch of Campaign Central, a new component of its Alerts and Notifications customer contact solutions. Campaign Central is a Web-based tool that allows

users to create, launch and manage their own outbound notifications to their customers via voice, SMS text message, email or fax. Through Campaign Central‘s Web interface, clients can take control of their customer contact solutions without additional costs that are usually associated with professional services fees. Users can also connect with their customers right from the desktop, designing or recording their own messages and deploying them instantly. Real-time campaign reports and statistics allow tracking of each campaign‘s progress so users can measure results and make immediate adjustments to meet business objectives and create meaningful customer interactions. Campaign Central offers auto throttling of outbound voice calls, which helps to streamline contact centers, maximize agent time, and increase customer contact rates by connecting customers to agents with minimal queue times.

Telephone Tools Applications/Products/Services/Partnerships/Financials

"IVONA for Developers" Portal rapidly gaining popularity among accessibility device manufacturers and application developers:

IVONA Text to Speech is enjoying tremendous uptake of its recently launched IVONA for Developers initiative within the accessibility marketplace. The initiative allows developers of all kinds to access the IVONA Speech Cloud or IVONA Software Development Kit (SDK) via the www.ivona.com/developer portal. Accessibility-focused organizations and companies using IVONA TTS include: - The Royal National Institute of Blind People (RNIB); - HIMS in its Braille Sense OnHand, the voice-enabled Braille notetaker series, with digital compass and GPS receiver built-in; - NPN and E-Lektor are using IVONA voices in DAISY books readers for; - Klango Network - the world's most popular social network dedicated to the blind voiced by IVONA; - Levelstar, a development and manufacturing company dedicated to creating tools that ―level the playing field for the visually impaired‖ plans to embed IVONA in a number of upcoming devices. Mobile accessibility devices voiced by IVONA Text-to-Speech Marc Mulcahy is co-founder, developer, and chief product engineer at LevelStar. Marc is also blind and had this to say about IVONA: “When it came time to select a text-to-speech partner for our next-generation products, IVONA TTS was the obvious choice. Not only are their voices extremely natural, but they perform well at the device level and are intelligible at high speeds, something that's very important to our users.” An example of a unique new accessibility project enabled by a single developer and the IVONA Developers Initiative is the Talking TV Guide. Created by Ian Sharpe, a visually impaired application developer who is also a champion member of the Great Britain Paralympics Team. Ian‘s Talking TV Guide allows users to hear and interact with TV listings on any computer or web-enabled mobile device

Page 19: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 19

so they can easily access program information and have a productive and pleasant learning experience. In his words: “Winning is all about commitment, passion and focus. This carries over into my work. My passion is to make the web easier to use and I think the Talking TV Guide reflects this.” More information on Talking TV Guide. ―IVONA for Developers‖ (www.ivona.com/developer), formally launched in February, is designed to offer everything a developer might need to incorporate text-to-speech functionality into an application or device. Through the portal, developers and manufacturers can access the IVONA Speech Cloud Service, which received in excess of 50 million hits in February Alone, and SDK downloads for Android, Windows 7 and Linux, with iOS, QNX and RIM/BlackBerry coming soon.

I6NET and IVONA Text-To-Speech Partner To Provide New TTS Technology and Voices For VXI* VoiceXML Browser on Asterisk Platforms: IVONA Text-to-Speech, global speech technology provider, and I6NET, provider of IP/3G telephony IVR/IVVR solutions for Asterisk, announced the availability of a new IVONA Text-to-Speech (TTS) connector to the VXI* VoiceXML browser for Asterisk platforms. With VXI*, customers can access IVONA high-quality TTS voices to send and receive voice and video messages, listen to phone podcasts, newspaper articles, etc., all leveraging IVONA‘s highly intelligible, human-like Text-to-Speech. The VXI* high-optimization integration with Asterisk provides advanced voice and video services, including protocol conversion, SSML, media conversion and streaming, application logic and third party integration. It supports industry standards for and is compatible with all devices supported by Asterisk. IVONA TTS voice are available in: British English, American English, Polish, German, Spanish (Castilian), American Spanish, Romanian, French, with many more coming soon, e.g. Italian.

Nu Echo & Loquendo Announce Partnership to Expand Professional Services Offering:

Nu Echo and Loquendo announced the launch of their new partnership, enabling the companies to offer North American customers speech application development, dialog design, porting and tuning services to complement Loquendo‘s existing portfolio of highest quality multilingual technologies, for a simpler yet more effective and enhanced integration of speech. This partnership reinforces the ongoing expansion of Loquendo‘s North American presence. Partnering with such an expertise-rich company as Nu Echo will see significant advantage for enterprises in the contact center and speech-enabled application space. It will expand the service level for Loquendo‘s partners in the U.S. and Canada, to include VoiceXML application development, grammar development for advanced speech recognition, dialogue and VUI design, TTS and ASR tuning, and so on.

VBVoice Integrates with Microsoft Speech Server through Media Resource Control Protocol (MCRP): Pronexus announced that

VBVoice developers can now take advantage of high-quality and cost-effective Microsoft Speech Server automatic speech recognition (ASR) engines with the integration of Media Resource Control Protocol (MRCP).

The MRCP Connector is a plug-in that allows VBVoice IVR developers to utilize cost-efficient Speech Server and text-to-

speech (TTS) products available in Microsoft‘s Office Communications Server (OCS). This announcement follows VBVoice‘s recent support for Nuance Recognizer 9. The MRCP Connector is provided by Aumtech. VBVoice is a free-of-charge rapid application development (RAD) IVR toolkit which enables the creation of powerful, speech-enabled solutions including voice self-service, unified communications systems.

Voxeo Adds Mobile Web Access to Cloud-based VoiceObjects On-Demand: Voxeo Corporation, announced that VoiceObjects

On-Demand, the new cloud-based version of its popular VoiceObjects Application Lifecycle Management Suite, has expanded to offer mobile web access as an additional customer

Page 20: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 20

communication channel. Customers can immediately deploy voice, text messaging, mobile web and Twitter applications into Voxeo's carrier-grade cloud infrastructure, backed by the industry's only 100% uptime guarantee. VoiceObjects is the complete Application Lifecycle Management Suite that provides the tools needed to engage customers in better, more efficient communications. Hundreds of enterprises and service providers have used VoiceObjects on their premises to realize measurable service improvements and quantifiable operational savings. VoiceObjects On-Demand makes the solution instantly available with no up-front investment. Developers use the well-known Eclipse environment to build, deploy, manage and refine applications. Development is free. Going live is seamless. With VoiceObjects On-Demand enterprises or service providers can: Accelerate development and dramatically simplify ongoing maintenance Improve the customer experience with actionable, real-time analytics across communication channels Improve loyalty and drive revenue with personalization Eliminate duplicate investments with a single solution that manages the entire application lifecycle across multiple communication channels Reduce support costs with a Unified Self-Service™ strategy that enables low-cost, text-based interactions Create and test applications faster and eliminate the time and expense of deploying hardware on your premises Speed migration from legacy IVR to open standard VoiceXML with VoiceObjects GUI development, automated migration tools, and the expertise of Voxeo's 24x7 Customer Obsession Teams VoiceObjects On-Demand additionally provides rich analytics to deliver deep, real-time insight into what customers like and dislike about self-service applications. This insight allows businesses to tune and optimize their self-service solutions to enhance the customer experience, increase customer loyalty, and improve automation rates. The service includes detailed reports on average call duration, business task completion rates, speech recognition quality, most frequent users, input analysis, customer usage paths, and customer exit points. Reports can be viewed live in Voxeo's web-based customer portal and can also be batch-exported into existing Business Intelligence platforms from IBM Cognos, MicroStrategy, SAP BusinessObjects and Oracle BI.

Mobile Applications/Products/Services/Partnerships/Financials

ZoomReader App Released for iPhone: Ai Squared announced the release of its iPhone application, ZoomReader. Designed specifically for anyone in need of visual assistance, ZoomReader takes a picture of any text, converts the text digitally using Optical Character Recognition (OCR), and reads the text aloud through an integrated voice synthesizer. The ZoomReader app can be used to read small print on a receipt, menu, box or medicine bottle, and is the first and only one of its kind on the App Store designed for the visually impaired that integrates live video magnification with up to 4X zoom, OCR and text-to-speech for the accessibility-friendly iPhone. ZoomReader features a simple user interface with large, prominent buttons, as well as customizable color filtering/highlighting, text size, and speech rate. The application also integrates iPhone‘s VoiceOver feature to ensure accurate screen control and speech commands for ease-of-use. Synthesized audio is available in English with the initial download, while three additional languages are available through in-app purchase. Available today, the ZoomReader app for iPhone may be downloaded from the App Store for $19.99

Play Antakshari on Mobile: Antakshari is one of the most popular musical games in India and as we grow to be a telecom generation, there is a need to expand the game across the virtual network. Mobile Antakshari is an interesting multilingual speech recognition technology mobile game that is based on the classic Indian music game of Antakshari. Developed by Chennai based Hexolabs (they earlier launched twitGuru – Human Powered Ad Network for Twitter), the product needs no install and is currently available as VAS on Idea network

Page 21: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 21

(currently Mobile Antakshari is available in 4 India languages – Hindi, Telugu, Tamil and Malayalam). The voice recognition part is powered by Nuance and beauty of the product remains simple usable interface that could be accessed by any number literate user.

Taskmind iPhone app from Catalysts: Taskmind from Catalysts GmbH is an organizer app for an iPhone or iPad, storing tasks, projects, and appointments. One can just speak a task, and turns the spoken words into text. Currently, German and US English are supported, with more languages planned. Other features do such things as distribute tasks to coworkers. The speech recognition is performed by the Nuance Mobile SDK through the Nuance Mobile Developer Program.

Cyberon Voice Speed Dial for Android; To help users of Android-powered mobile devices conveniently make phone calls using only the power of their voice, leading embedded speech solution provider Cyberon Corporation announces its latest mobile application, Cyberon Voice Speed Dial. The application allows users to make phone calls to friends, family, and colleagues by speaking their name. The highly accurate speech recognition technology quickly dials their phone number, eliminating the need to hunt through an address book or remember a phone number. The application also connects with a user's Bluetooth headset (for Android 2.2 or above) to make calling safer when driving. Features of the Cyberon Voice Speed Dial application for Android include: Make phone calls by speaking a name - The application listens to what users say for quick and easy dialing Make calling while driving safer - Drivers can avoid distraction with this hands-free solution Highly accurate voice recognition - Because it is based on a user's actual speech, the application can quickly and accurately understand their voice Works with Bluetooth headsets - Users can dial and talk hands-free by connecting a wireless headset Removes language barriers - The application can recognize virtually any voice or language Unlike other voice recognition solutions that require the user to spend large amounts of time doing training exercises, Cyberon Voice Speed Dial can start working more quickly. Users only need to create a "voice tag" for each person they would like to call using the application, which allows for easy setup and highly accurate speech recognition. This "voice tag" technology also allows the application to recognize virtually any voice, without language barriers. The display can be changed to any of 20 languages, including English, German, Spanish, French, Italian, Simplified Chinese, Traditional Chinese, and more. Cyberon Corporation developed the application using their experience from over ten years of creating voice recognition solutions. Other mobile applications created by the company include Cyberon Voice Commander, Cyberon Talking Dictionary, and Cyberon eEnter. Cyberon Voice Speed Dial is now available worldwide for $2.99 USD through the Android Market (under the Tools category). The application is compatible with Android-powered mobile phones version 1.6 or above Dial2Do Powering SuperTooth HandsFree Assistant: Dial2Do has been chosen by Bluetooth speakerphone manufacturer SuperTooth to power a new service called SuperTooth HandsFree Assistant that will enable users of the new SuperTooth HD speakerphone to have two-way, voice-activated, hands-free messaging. As part of the white-label arrangement, SuperTooth HandsFree Assistant users can use their voices to create and send SMS, email, and Twitter and Facebook posts and also have incoming SMS and email read aloud on Android and BlackBerry phones via the specific free application. New SuperTooth HD users receive six months of the SuperTooth HandsFree Assistant service. During this period, the user can upgrade to the SuperTooth HandsFree Asistant Pro service for one year, which includes a range of additional features, such as the ability to listen to local news, sports, and weather. At the end of the six months, users can still upgrade to the Pro service or they can subscribe to continue using SuperTooth HandsFree Assistant Basic.

Page 22: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 22

Epocrates to add speech recognition to its mobile EMR: Epocrates is testing an integrated mobile dictation feature, called ―SpeechAnywhere,‖ that will allow physicians to create medical reports on the fly, via their smartphones. It's a crucial link-up to give physicians the ability to not only access and read their patient records on their mobile devices, but also to input information. The fastest input method for a doctor on the go is dictation. "When you have the ability to dictate, you no longer need a keyboard — you‘re fully mobile," says iMedical apps reviewer Iltifat Hussein. And that's where speech recognition comes in -- translating the doctor's dictation directly into the record and bypassing the usual dictation/transcription/review cycle -- which can slow things down by days, Google Voice Search now available in Bahasa Malaysia:

Google officially announced the availability of its Voice Search in Bahasa Malaysia, joining the 19 other languages already available for the utility. It offers users a fast, natural way to search the web from their mobile phones by speaking the queries instead of typing them. Using Google‘s mobile app, users can search on the go by simply clicking the microphone button beside the search bar and speaking into the phone in BM (or any of the other languages that Voice Search recognises). It‘s available for Android 2.1, 2.2 and 2.3, iPhone and Blackberry. If you‘re running Symbian, you‘re out of luck. On Android and iPhone mobiles, all you have to do is tap the VS icon and begin speaking, while on Blackberry, you press and hold the green call button to get things going. It‘s versatile – you can browse for photos, look for places of interest, search for an address via maps. The likes of ―Gambar Sazzy Falak‖ and ―nasi kandar Jalan Tuanku Abdul Rahman‖ gets you search results on your mobile browser, while on maps the name of the street gets you the desired

location in a jiffy. There‘s no need to download specialised speech recognition software in order to use Voice Search – queries are processed in a cloud using cellular Internet connections, and not on an individual device. The neat part is that the more people use VS, the better the speech recognition models get, improving accuracy without the need for updates or the installation of new software. The recognition factor is high, despite the diversity of spoken and written forms of BM and the different regional accents, as well as normal mix of English and Malay used in everyday phrases (pergi meeting, for example). In the car, it‘s quite the helpful hand. To build Voice Search, Google utilises computers to understand the sounds and words that make up a spoken language by working with native speakers to collect speech samples. In the case of the BM version, it was developed with the help of 500 local student volunteers, of which their speech samples were then integrated into the language models powering the service.

Sprint Now Offers Google Voice Apps on All Its Phones: Sprint and Google announced a new integrated Google Voice experience that will allow Sprint customers to set their existing Sprint wireless phone number as their Google Voice number. Sprint customers will be able to take advantage of all the features and benefits of Google Voice, including call recording and voicemail transcription, without the hassle and fees associated with porting their numbers. Sprint is the first and only carrier to offer this capability, which will be available soon on all Sprint CDMA phones. In addition, the new Nexus S 4G phone, announced today by Sprint, Google, and Samsung, will come preloaded with the Google Voice Android app. Google Voice lets users manage up to six different devices through one phone number with intelligent call routing and advanced features like call screening, blocking, and recording using the phone‘s dial pad. Sprint subscribers will be able to use their Sprint numbers across their office, home, and wireless phones, and personalize settings so calls from friends ring their wireless devices and home phones,

Page 23: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 23

while calls from the boss only ring at the office. Google Voice users can receive transcribed voicemails and read or listen to them online. They can also read, send and search text messages and call logs online

NHN and Hyundai-Kia Band Together to Develop Telematics Services: The Hyundai-Kia Automotive Group, the leading automaker of Korea, joins forces with Naver, the biggest Internet portal in Korea for development of telematics services. The automobile group will offer platforms while NHN, operator of Naver, various contents. NHN (CEO Kim Sang-heon) and Hyundai-Kia Automotive Group (Chairman Chung Mong-ku) announced on April 10 that the two signed an MOU on building partnership in the next-generation IT / telematics service sectors and would cooperate for development of automobile-specialized smart services and popularization. The telematics terminal to be developed jointly will use Naver‘s proprietary voice recognition technology. This technology will enable drivers to search various information such as news, weather, music and stocks. In addition, the driver can listen to the music streamed from Naver and use navigation service to search good restaurants and famous places, as well.

iSpeech Releases DriveSafe.ly on iPhone: iSpeech has launched DriveSafe.ly on iPhone, the first iOS app to speak text (SMS) messages, emails and Tweets aloud and allow for text responses by voice. The free app has been used by more than 8 million people and has spoken nearly 500 million messages to users. iSpeech technology powers DriveSafe.ly and other apps with text-to-speech (TTS) and speech recognition (ASR). Once activated, it is possible to message via TTS and ASR without touching the phone. Once an incoming text message is read aloud, the user can choose to replay the message, speak a response back to the sender, or even call back the sender. DriveSafe.ly is available for iPhone in the iTunes App Store, BlackBerry in the BlackBerry App World, and Android in the Android Market.

Warid Congo launches Voice SMS on Kirusa platform: Warid Congo SA, the third largest telecommunication operator in the Republic of Congo, announced the launch of Voice SMS across its network. The offering is based on the Voice SMS technology from Kirusa. Voice SMS allows Warid Congo customers to send messages to any mobile subscriber in the Republic of Congo by voice, eliminating the need to type. The ease of use of the technology also allows mobile subscribers to experience the emotion and convenience absent in text messaging. Use of Kirusa Voice SMS by mobile phone users is gaining global traction due to its ease of use and value provided to mobile subscribers.

MobiFlex for Google Apps: MobiFlex, the online service to create native mobile apps, is now available through the Google Apps Marketplace, Google's online storefront for Google Apps products and services serving over 3 million businesses. Now, from this central online location, millions of Google Apps companies can access and use MobiFlex to create robust native mobile business apps for Android phones and iPhones directly integrated with Google Docs Spreadsheets for Backend Data without coding - at 1/10th the cost, in 1/10th the time of comparable custom developed apps.

Carolina West Selects Movius Voice-to-Text Solution: Movius Interactive, a provider of unified communications solutions, announced that Carolina West Wireless has selected its Voice-to-Text solution. Movius‘ Voice-to-Text solution has been developed jointly with Ditech Networks to provide carriers with a contemporary offer for voice messaging. It uses Movius‘ Versera ICE Value Added Services platform and Ditech‘s PhoneTag voice-to-text solution, giving mobile subscribers a way to read voice messages as text on their handsets.

Page 24: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 24

Nuance Dragon Dictation and Dragon Search apps in the China iTunes store: Nuance announced that the Dragon Dictation and Dragon Search applications for iPhone, iPad, and iPod touch are now available in the China App Store with support for Mandarin Chinese. The Dragon Apps have been

rapidly expanding in recent months to give even more consumers worldwide the ability to create emails, text messages, status updates, and search the web simply by using their voice. The apps are now available in 11 different languages – US, UK and Australian English, European and Canadian French, European and American Spanish, German, Italian, Japanese, and Mandarin. Dragon Search for iPhone and iPod touch lets users speak any search query to find any content on the mobile Web. Nuance's Dragon Search Carousel delivers search results within seconds across multiple web content sources including Google, Sina, Dianping, Youku, and taoboa.

Nuance Powers Movistar’s Llamadas Perdidas Voice-to-Text Service: Nuance announced that its Missed Call Messenger voice-to-text platform now powers Movistar Spain‘s Llamadas Perdidas, a suite of voice-to-text services that enables its full subscriber base to receive transcribed voice

messages via text message when call recipients can‘t be reached or do not have voicemail. Movistar‘s Llamadas Perdidas completes the call every time, providing subscribers with the ability to speak a message up to 15-seconds long that is transcribed and delivered via text message along with the number of the caller to the recipient. Movistar‘s Llamadas Perdidas is powered by Nuance‘s Missed Call Messenger service, which is built on Nuance‘s voice-to-text platform, currently deployed with mobile operators around the world. The service is automatically available to all Movistar Spain prepaid and contract customers, with no charges incurred by the message recipient.

Parrot Asteroid Brings Internet To The Car: Parrot claims it‘s Asteroid receiver will be ―connecting the car to the internet.‖ The Asteroid will offer connection to thousands of internet radio stations and podcasts via an Orange Liveradio app. There will be three other apps bundled with the Asteroid. Maps will offer GPS localisation but no turn-by-turn navigation. Based on Google Maps, the app will give you details of local points of interest, text-based directions to those places as well as phone numbers and other details.

The final two apps which will come with the device will be subscription-based. The first of which will be Fuel for Less, which will let you know where service stations near your location are and where the cheapest fuel is to be had. iCoyote on the other hand will give drivers live updates of traffic jams and accidents as well as alerts on up-coming speed cameras. Parrot believe the this is ―going to be a very open platform with new apps coming in the next couple of months.‖ They confirmed to us they were working with other developers at the moment. The Asteroid will connect via Bluetooth to practically any phone and has six different phonebook sync protocols loaded up to allow for most phone OSes to sync with it. Hands-free calling is facilitated via the voice recognition button on the left of the receiver. The voice recognition software does not have to be trained and automatically recognises accents as well as seven languages including English, French, Spanish and Portuguese. The 1 DIN standard-format receiver is split into two parts. On the right is a 3.2in TFT screen while the left is taken up with a detachable faceplate on which are back-lit buttons including some familiar Android buttons and the dedicated voice recognition button. This will also allow you to search your music by artist, album or track name. Music can be stored on an SD card, with the slot located under the faceplate or via USB on a PMP. The receiver has four USB ports at the rear - one for the included GPS antenna, one for a 3G dongle and two free for your PMP or other USB device. Included in the pack are USB cables to allow you bring the USB connection from the rear of the receiver to your glovebox or central console. 3G tethering via Bluetooth is currently possible and a software update will bring USB tethering in the near future, Parrot said at the launch today.

Page 25: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 25

Quickoffice Unveils Custom-Built Mobile Office Suite for Latest Android Tablet Devices: Quickoffice announced a newly designed application for Google's Android Honeycomb OS, catering to the tablet's larger form factor and touch user interface. Called Quickoffice Pro HD, the mobile Office suite delivers desktop-like creating, viewing and editing capabilities of Microsoft® Word, Excel and PowerPoint files. Utilizing both Quickoffice's proprietary SmartTouch technology and the unique capabilities inherent in Google's Honeycomb platform, the app offers a compelling single-screen, multi-pane document view for a richer user experience on tablets. In addition to feature-rich Office viewing and editing capabilities, Quickoffice Pro HD comes complete with a proprietary and custom-built Connected File Manager with integrated access to cloud storage partners, including Box.net, Dropbox, Google Docs, Huddle, SugarSync and MobileMe, all within an intuitive file browser. The newly designed user interface leverages native Honeycomb elements, like an action bar, contextual toolbar, fragments, text-to-speech and dictation with speech input functionality. The action bar provides users with quick access to common functions, contextual options and navigation. The contextual toolbar changes to reflect options based on the context of what the user is currently doing. Quickoffice also integrated Android's fragments to enable users to interact with multiple panes independently of each other, all within a single screen view. Quickoffice Pro HD offers three powerful editors within one easy-to-use application, including: Quickword — page layout editing; quick page scrolling to jump to pages in large documents; insert/resize/rotate images from camera and gallery; text-to-speech; undo/redo; and dictation with speech input Quicksheet — formula builder; edit directly in cells; add/remove/rename worksheets Quickpoint — format text; create new presentations; insert/remove slides; full screen mode with manual flick scroll and automatic change of slides; rearrange slides; and Beyond integrated cloud access, tablet users also have the option to access and share files via email or Bluetooth. Within the file browser, consumers can drag and drop to send, delete or move files. Quickoffice also supports local and remote search of files with its integrated Quicksearch function to help increase productivity on the go. Consumers can also leverage the tablet's HDMI output to present presentations when a larger screen is needed. Quickoffice comes pre-loaded on Motorola's XOOM tablet and consumers have the option to upgrade for full editing and cloud access functionality for an introductory price of $14.99.

SoundHound music recognition software adds recommended songs: SoundHound announced an update to its music search and discovery application: Recommended Songs. The new feature will be available exclusively to users of SoundHound's paid offering, SoundHound Infinity. The feature recommends top picks for songs users might enjoy based on the artist or song they are checking out. It is now available for download on iOS devices as part of the new SoundHound version, 3.6.1. The Recommended Songs feature offers easy, instant suggestions. When using SoundHound's music recognition – renowned for working even if a user sings or hums – users will see a "Recommended Songs" link on a result pages. Alternatively, users can get instant access to Recommended Songs just by speaking a song title or artist name into SoundHound's instant voice search. SoundHound generates Recommended Songs through a proprietary engine that weights SoundHound's Similar Artist, Underplayed and Hottest data. Song exploration has never been so easy and delightful.

SpeechCycle Launches SmartCare Mobile: SpeechCycle launched SmartCare Mobile, a smartphone platform and application suite for customer care. The product launch comes one week after SpeechCycle announced its creation of a mobile division to design applications specifically for smartphones. Companies can plug SmartCare Mobile into their existing customer service applications to create custom widgets for customer surveys and feedback, notifications, billing, support, and scheduling. SmartCare Mobile also features an integrated mapping feature that allows companies to send geographically targeted advertising messages to interested customers. SmartCare Mobile supports all of the leading global smartphone operating systems, including Android, iOS, Windows Phone 7, and BlackBerry, on a single application platform. Key features of SmartCare Mobile include: intuitive navigation with T3 (Touch/Type/Talk) interface; predefined, customizable widgets;

Page 26: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 26

a dynamic deployment model; and seamless integration with enterprise contact centers and back-end systems. It consists of a native device client, cloud-based services, and an extensible integration framework that ensures the user interface and interaction experiences are optimized for smartphones. It doesn‘t require the user to go to a company‘s Web site, but uses an embedded browser to allow customers to access the Web if needed. SmartCare Mobile is a white-label application that customers can access when they download a paricipating company‘s customer care application. Several companies are now piloting the application. The goal of SmartCare Mobile is ―to take calls out of the contact center and optimize the customer experience,‖ Pan says.

TeleNav hybrid navigation supports connected and on-device routing TeleNav announced the launch of its hybrid navigation technology. TeleNav‘s technology provides premium turn-by-turn navigation and local search capabilities by either using real-time data sent to the phone over the wireless network or data that has been cached on the device. Applications incorporating the technology will automatically switch between the two navigation methods as needed to ensure users are never without access to local search and spoken turn-by-turn directions, whether they are in or out of wireless coverage. The dual-navigation capability is a highly requested feature among mobile GPS navigation users* and TeleNav meets this request by combining the benefits of both connected, over-the-network navigation with on-device, cached navigation data. Over-the-network navigation gives users access to advanced features and real-time information to deliver an enhanced personal navigation experience. On-device navigation allows users to search for and get turn-by-turn directions to businesses, local landmarks, or addresses while not connected to the wireless network. The over-the-network navigation capability of TeleNav‘s hybrid technology enables features such as real-time traffic information, automatically updated maps, satellite view provided by Bing™ Maps, one-touch voice search, lane assist, and access to search more than 22 million local Places in the U.S., including real-time local gas prices. Applications featuring the technology will use over-the-network navigation as the default navigation method and only switch to use the content cached on the device when necessary. Once a user has installed a GPS navigation application featuring TeleNav‘s hybrid technology, the application will automatically switch to use cached navigation data in instances of wireless network interruption. This on-device navigation allows users to receive turn-by-turn directions and conduct local searches even when they have no wireless data coverage. Other features available include spoken turn-by-turn directions with street names, 2D and 3D moving maps, automatic rerouting when a turn is missed, and the option to choose route preferences, such as avoiding toll roads.

TomTom Selects VoiceBox for Latest GO and VIA Products: VoiceBox Technologies announced that it was selected by TomTom as a provider of voice technology in the US and Canada. TomTom recently launched its new GO and VIA Series in the US with VoiceBox functionality. VoiceBox is working to alleviate the ‗cognitive load‘ for drivers – taking your mind off what you‘re doing while driving – a safety concern which received increased attention at the US Department of Transportation‘s 2010 Distracted Driver Summit. Both the newly launched GO (GO 2435, GO 2535 and GO LIVE 2535) and VIA Series use VoiceBox technology to bring premium safety-enhancing features that help address increasing distracted driving concerns.

VoiceBox has partnered with SVOX AG to deliver a better address-entry experience that utilizes SVOX‘s speech recognition engine. Now, when using VoiceBox‘s Conversational Voice technology, TomTom device users can speak naturally; so instead of speaking an address one step at a time, they can simply

say, ―Take me to 1313 Mockingbird Lane, Seattle.‖ The user can speak to the TomTom device with natural number forms like, ―thirteen, thirteen‖ instead of having to translate them into un-natural forms, such as ―one-three-one-three‖, forced by the limitations of traditional voice recognition technology. With the VoiceBox solution, TomTom device users can quickly and more naturally utilize voice commands, such as entering an address, further helping them keep their eyes on the road and hands on the wheel.

Page 27: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 27

Vlingo Launches Its Virtual Assistant in the UK: Vlingo announced English language localization of Vlingo for its UK users on all Android phones. By combining voice recognition, natural language processing and Vlingo's proprietary Intent Engine, the Vlingo Virtual Assistant understands

the user's intent and takes the desired action. Users simply speak to their phone to send text and email messages, search the web, get directions on Google maps, update Facebook or Twitter status, and more. Vlingo helps users to do just about anything while on the go.

Voice Assist Partners with Jabra to Provide Complete Hands-Free Solutions: Voice Assist and Jabra have joined forces to deliver complete end-to-end, hands-free solutions for consumers and enterprise users alike. With the combination of the Voice Assist cloud-based speech platform and services with Jabra Bluetooth headsets and in-car speakerphones, users can increase productivity using any mobile phone or IP-connected device (such as the iPhone, Blackberry, Android phones, the iPad and other tablets) at work, at home, on the road, and on the roam. Additionally, when paired with Jabra's speakerphones, like the upcoming Jabra FREEWAY, the combination provides a hands-free driving solution, enabling the user to make calls, send or respond to text messages and emails, or post to social networks such as Facebook and Twitter, all by voice. Users simply press one button on the Jabra headset or speakerphone, and then speak as they normally would, and say what they would like to do. Through this partnership, both companies will be marketing each other's products either separately or as a bundled solution. Bundled products will be offered through multiple direct and indirect sales channels.

VoiZapp, Inc. Releases Friends Aloud iPhone app: Apple iOS device owners can now access and listen to their Facebook news feed via Friends Aloud, created by VoiZapp, Inc. It is the first iOS app to bring optimized state of the art text-to-speech technology to Facebook on all iOS devices. Using Friends Aloud, a user can: · Read aloud their latest Facebook news feed posts and their associated comments in a pleasant, easy-to-understand young woman‘s voice · Scroll through each post on screen or tap around to read posts at random · Automatically switch between reading existing posts in sequence or just reading new posts as they arrive for as long as desired · Play, pause, and skip forward/back using intuitive controls, just like the iPod · Continue reading aloud while in the background, just like the iPod · Understand and read common computer jargon such as smiley faces and texting-style abbreviations (LOL, ROTFLMAO, happy/sad face, etc.) · Choose to limit the number of comments read aloud

Weather Radio App automatically alerts of life-threatening weather: iMapWeather Radio from Weather Decision Technologies, Inc. (WDT) is a mobile application for Apple or Android devices that sends alerts to the user when weather danger approaches. It also offers instant voice and text alerts which are based on location preferences. Voice alerts come from the National Weather Service during severe weather events.

Yahoo search for iPhone and iPod touch: Yahoo! announced the release of Yahoo! Search for the iPhone and iPod touch. Yahoo! Search, is a standard search application offering voice search, local results, maps, and personalized results. The Yahoo! Search iPhone app delivers contextually relevant results quickly. Features include voice search, tightly embedded maps, localized query suggestions that display as one types, intelligent search history, and rich content and information from Yahoo! and other sources that appear directly in the results Yahoo! Search is currently available in 22 countries.

Page 28: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 28

Yap now offering Spanish voice-to-text services: Yap has launched its transcription service for Spanish speakers. ―In just the past six months alone, we‘ve seen the volume on our platform spike dramatically as a range of businesses realize the power of Yap‘s speech recognition platform,‖ says Jablokov. ―We‘ve been applying the lessons learned from English

language development to allow us to offer high quality automated transcriptions of Spanish voice input. Given there are more Spanish speakers in the U.S. than in Spain, it‘s a

significant market.‖

Yap to Provide Free Voicemail-to-Text for Android: Yap announced the availability of Yap Voicemail for Android. Yap's goal is to eliminate the need to listen to long voicemails by converting voice messages into text presented in a clean, organized manner.

The application's key feature is the ability to quickly find and read voice messages. Additionally, Yap Voicemail allows its users to respond to voice messages via text

message or email. In an attempt to provide its services directly to consumers, Yap Voicemail is a free download, can be found in the Android Market. Yap Voicemail converts voice messages into text and presents the results in an intuitive graphical user interface on their mobile devices. Key features of Yap Voicemail include: Quickly Read Voicemail Messages: Yap Voicemail lets users glance at their Android phones to quickly preview messages and determine whether they require immediate attention. Find Important Messages Quickly: Instead of listening to each message until finding that critical piece of information, Yap Voicemail lets users type in a few keywords that describe what they are searching. Yap Voicemail displays transcriptions of all of the voice messages that match. Reply Using Email and Text Messages: Just because someone called and left a message doesn‘t mean a return phone call is required. Thanks to reply functionality that is integrated with Android‘s native SMS and email, Yap Voicemail lets users conveniently and quickly respond to voicemail messages via a text message or email.

ZoomSafer closes additional $1.1 Million financing: ZoomSafer announced that it has raised an additional $1.1 million round of financing from White Birch Capital and SugarOak Holdings. Proceeds will fund expanded marketing and further development of the company‘s software products.

Embedded Applications/Products/Services/Partnerships/Financials

Bark recognition: The Bark Terminator 2 collar for controlling a dog‘s barking made by High Tech Pet Products uses ―voice recognition‖ to reduce false alarms. The Terminator 2 first applies a warning tone followed by progressively increasing tone and shock as your dog‘s barking persists. If your dog is quiet for 30 seconds or more, the Terminator 2 resets to the mildest correction level. In this way, the pet needs only to experience the higher shock levels a few times before learning that the mild sound and shock stimulus is a warning of stronger corrections to come.

Audio system for aircraft from Garmin uses voice commands: Garmin International announced the GMA 350 and GMA 350H audio panels for fixed wing and rotorcraft, respectively. The GMA 350 is a digital audio panel with marker beacon that has a new speech recognition feature so that pilots can use their voice to control certain aspects of the audio system. For example, pilots can select COM 1 or COM 2 by simply stating ―COM 1‖ or ―COM 2‖ rather than pressing a key. Voice commands can be used to control most aspects of the audio panel that would normally be done with a button or knob. The GMA 350 also includes 3D spatial processing, which mimics the way people process sounds and conversations. Humans have the ability to focus their listening attention on a single talker among a mixture of other conversations and background noises, ignoring other conversations. Garmin has

Page 29: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 29

leveraged this ability by simulating 3D sound with stereo headsets. For example, if the pilot is monitoring air traffic control on one radio and the weather, such as ATIS, on another radio, the GMA 350 will separate the two conversations by placing air traffic control towards the pilot‘s left and the weather towards the pilot‘s right. This enhances the pilot‘s ability to understand conversations that are happening simultaneously.

Loquendo Selected as Speech Synthesis Provider for SICOMBUS Multimedia Transport Information System:

Sicombus allows passengers to receive travel information in real time on city buses routes. This solution has been developed by two Spanish companies, Proconsi and Icon Multimedia, integrating Loquendo‘s Text-to-Speech. With an R&D investment of €400,000, this Digital Signage Project will encourage the use of public transport, increase passenger confidence in the service, and transform the daily routine of millions of users into a positive experience – so changing society‘s perceptions about the value of public transport. Among the many options available, the solution offers travellers useful route information: bus timetables, stops and routes, connections with other lines and other transport networks. SISCOMBUS is also able to manage other types of information such as news, weather forecasts, tourist information, plus LBS and what‘s on information for each point on the journey. Compared with other existing Digital Signage systems, Sicombus offers key improvements such as the use of dynamic information based on the GPS position of the bus, enabling the creation of personalized publicity channels and the enhanced provision of local cultural information. The system also delivers connection information with other transportation systems such as trains, flights, etc, so delivering a multi transport information channel with real benefits for passengers. Information is updated in real time using a central Text to Speech Server powered by Loquendo‘s accurate and natural sounding synthetic speech – available in various languages and voices. No pre-recorded audio is used, enabling messages to be created dynamically from a variety of sources and third parties systems, including advertising, and thus continually updated.

Loquendo Morphs the Voice of the ‘The Gadget Show’ Presenter for UK’s Channel 5: Loquendo announces its participation in the ‗Gadget

Show Challenge‘ – to create a cloned voice and robot of one of the real life presenters of the UK‘s Gadget Show, Jason Bradbury. Loquendo created an artificial version of Jason‘s voice by morphing the voice of Simon – one of Loquendo‘s British English synthetic voices. The Gadget Show Challenge was to allow two of the presenters to host the programme using robot and avatar versions of themselves. As part of this challenge, many experts and organizations collaborated to create a robotic version of Jason, known as J-Bot, with the synthetic speech technology of Loquendo. UK residents can watch the show here. To transform the source voice into the target one, a comparison was made of the two voices using recordings of the real voice of Jason and the synthetic voice of Simon. Features such as fundamental frequency and formant positions were analysed. The pitch of Simon‘s voice was adjusted to bring it closer to Jason‘s, and the positions of frequency resonances were tuned. Digital filtering was also performed to even out spectral energy distribution.

Microsoft Kinect achieves record sales rate: Microsoft said it has sold more than 10 million Kinect motion-sensing game system units with voice control worldwide in just over four months, the fastest-selling consumer device on record. Microsoft said the popularity of the Kinect had also spurred the sale of more than 10 million stand-alone Kinect games.

NICE Recording Technology to Bolster Security at DFW Airport: Dallas/Fort Worth International Airport (DFW) has placed a multimillion dollar order to expand its NICE implementation. As part of this initiative, the airport will also be expanding its audio recording

Page 30: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 30

capabilities, adding the NICE Inform multimedia solution, which enables authentic incident reconstruction through synchronization of voice recordings and video. When the project is complete, DFW will be able to centrally monitor, record, and manage all IP and analog cameras seamlessly in one system.DFW will also be upgrading its recording and incident management capabilities with the NICE Inform multimedia incident information management solution. The addition of NICE Inform means that DFW, for the first time, will be able to synchronize voice recordings and surveillance video for authentic, 360-degree incident reconstruction. The NICE Security Offering enables capturing, analysis, and correlation of data from multiple sensors and systems, including audio, video, radio, geo-location and Web, providing a framework for fusing data silos into a single, holistic operational view

Desktop Applications/Products/Services/Partnerships/Financials

Babylon 9 Will Speak Acapela: Babylon, a provider of language solutions, has chosen Acapela Group to to speech-enable its Babylon 9 translation

solution, which provides one-click translation for any computer application, such as email, Internet pages, PDF documents, and Microsoft Word files. Acapela text-to-speech will support millions of Babylon users worldwide with audio pronunciation of a searched word, to optimize the translation process and language learning. Acapela TTS will offer an audio version of the translation with male and female voices in any of 18 languages. Other features include the Babylon Live Translation Community, a forum for translators to ask and answer questions and offer advice on translation, grammar and spelling, including current slang and dialects. Spelling and Proofreading are available. Babylon 9 also offers more languages, more sources, the ability to translate full Web pages and documents, compatibility with Microsoft Office Speller, and better access to dictionaries and encyclopedias.

Acapela Group Provides Voices for E-Doceo Online E-Learning Service: E-Doceo, a French producer of e-learning software, has contracted with Acapela to provide the voices for characters

in its online media store. The Acapela voices are used in conjunction with avatars from Cantoche to completely animate the Web portal. e-doceo media store enables the ordering, in a few clicks, of customized multimedia resources via an online platform. Files vocalized by Acapela voices and 3D animation are immediately downloadable from the customer's account. Actors' voices, illustrations, videos, and other animations are available for download a few days after the order. e-doceo media store relies on Acapela Voice as a Service (Acapela VaaS) to vocalize written content in real time. Technically, Acapela VaaS is a cloud computing solution that speech-empowers all Web applications. e-doceo connects to the Acapela VaaS server, sends the information to be vocalized, and the server automatically transforms the text into speech. This service is available around the clock and responds very rapidly, completing the process in several seconds. Speech synthesis enables learning units, which need to be developed in several languages or updated regularly, to be quickly animated with natural sound. All written information relating to the e-learning course is converted into voice, word for word.

ClearOne Unveils New Chat Speakerphones for Microsoft Lync and Skype at Industry Conference on Enterprise Communications

ClearOne unveiled its two latest USB speakerphones designed for personal desktop use in organizations using Microsoft Lync or Skype for audio conferencing via Windows personal computers. In Orlando for the opening the Enterprise Connect industry conference, ClearOne introduced the CHAT 70-U (for Lync) and CHAT® 60-U (for Skype), both featuring integrated call controls, including an end-call button, on the device. The new models are also equipped with propriety ClearOne technologies enabling natural conversation without the clipping, echo, noise or tinny sound many people associate with speakerphone use.

Page 31: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 31

Both new CHAT speakerphones, created expressly for VoIP applications using PCs or Unified Communications software, are designed for ease of use and superior voice quality. They feature raised, push-button controls for speaker volume, mute, and call hook and include a blue LED, which lights when a call is in progress. Just slightly larger than a typical computer mouse, the CHAT 70-U and CHAT 60-U incorporate ClearOne's HDConference technologies, which provide acoustic echo cancellation, noise cancellation, and automatic audio level control, all critical to allowing natural, free-flowing conversation and producing clear, rich sound quality. NoteSwift debuts:

NoteSwift debuts to the health IT marketplace with the help of 1450, Inc. With NoteSwift and Dragon Medical together physicians can input significant data into their EHR system with narrative speech. NoteSwift recognizes meaningful data in dictations, and places them in the EHR. Currently NoteSwift supports dictation into Allscripts Professional EHR versions 7, 8 and 9.0, and the company plans to release versions for many more EHRs in the future. A version of NoteSwift for Allscripts MyWay is expected to be released in the first half of 2011. Physicians using Allscripts Professional with Dragon Medical and NoteSwift have an ability that very few others in the industry have: the ability to get discrete structured data into their EHR from the spoken voice. A doctor simply dictates and together NoteSwift and Dragon Medical will take the relevant data from the dictation and place it into the appropriate data fields in Allscripts Professional. Based upon an exclusive distribution agreement between 1450 and NoteSwift, 1450 will be responsible for all distribution of NoteSwift's Electronic Health Record (EHR) / Electronic Medical Record (EMR) related programs.

Nattiq Adds Acapela's Arabic Voices to Dolphin Pen's New Generation: Acapela's Arabic voices Leila, Salma, Mehdi, and Saieda, as well as Heather, an U.S. English voice, are now speech-enabling the Dolphin Pen to ease daily life of Nattiq's visually impaired users, allowing them to bring Arabic speech capability as well as magnifying and Braille with them, anywhere. The new second-generation Dolphin Pen is immediately available with Hal Screen Reader and SuperNova Reader Magnifier versions 11.58. Acapela's set of Arabic voices, which are part of wide multilingual family of 30 languages and more than 60 voices, offers choice to the end user. A full set of natural voices are now available to Nattiq's users to select their preferred gender and persona.

Hamilton Relay Selected New Provider of Captioned Telephone Services in Virginia: Hamilton Relay announced that they have been selected by the Virginia Department for the Deaf and Hard of Hearing (VDDHH) to provide Captioned Telephone (CapTel®) services in the Commonwealth of Virginia. The service makes it possible for individuals who are deaf or hard of hearing to make and receive telephone calls using captioned telephone technology, ensuring call clarity for both parties. Hamilton began providing Captioned Telephone Service in Virginia on April 1, 2011.

Kopin Announces FCC Certification and Availability of Golden-i 3.5 Computing Headsets: Kopin Corporation announced Development Kits for its Golden-i 3.5 Hands-Free Computing Headset

are available to qualified customers at www.mygoldeni.com. The Golden-i 3.5 headset has been certified by the FCC. Delivering near perfect multi-language speech recognition with a virtual 15-inch PC display, the Golden-i headset provides immediate hands-free access to nearly any digital

information in the world and the ability to remotely control most wireless devices. Delivering near perfect multi-language speech recognition with a virtual 15-inch PC display, the Golden-i headset provides immediate hands-free access to nearly any digital information in the world and the ability to remotely control most wireless devices. The Golden-i headset is a light wearable mobile computing and communication system that provides wireless, hands-free control of up to six devices at one time (Smart Phones, PCs, Industrial Equipment, Servers and Cloud Environments), wherever

Page 32: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 32

there are wireless resources (e.g., Bluetooth, WiFi and/or Cellular). Golden-i Gen 3.5 Headset Development Kits are $2,500 each (plus tax and shipping) The Development Kit is designed for ease of use in adding speech recognition and gesture interface to industry standard and legacy software applications already running on PCs and enterprise systems. Each Development Kit includes: one Golden-i Gen 3.5 Headset; a Windows 7 application software emulation environment; Microsoft WinCE 6 application software porting platform; Nuance VoCon3200 Golden-i speech recognition platform available in more than 20 languages; online professional application development support provided by Adeneo, a Microsoft Gold Developer Partner; access to the global Golden-i users group; a second Li-ion battery and wall-plug battery charger; a user's guide; and a developers instruction manual.

Google adds speech recognition to Chrome beta: Google is updating its Chrome web browser to add voice recognition. The latest build of Chrome beta features support for the HTML5 speech input API, which Google has been working on with the HTML Speech Incubator Group. The API will allow developers to write web apps for Chrome in HTML5 that allow full speech recognition at the click of a button. The transcription occurs on a speech server after the sound is recorded. Of course, the nature of speech recognition software means that the faster you say something and the more of an accent you have the less likely it is to pick it up accurately. We tried the last phrase again in a slow monotone voice, which it picked up accurately. This might satisfy some, but we still dream of a day when we can talk normally at a reasonable pace while the computer picks it all up completely accurately. Google is also adding a GPU-accelerated 3D CSS feature, which allows developers to render 3D effects in CSS on web pages.

IVONA Text to Speech in The Talking TV Guide: The Talking TV Guide has been specifically designed by and for the visually-impaired. The new Guide combines customized content with web-based speech technology (IVONA Speech Cloud) to provide a highly efficient audio interface. This allows users to hear and freely interact with TV listings on any computer or web-enabled mobile device. Hot keys provide rapid navigation and instant access to further program information while the user listens, in real-time. High-quality IVONA TTS voices ensure The Talking TV Guide is not only the most efficient, but also the most intuitive programming guide available today. TV Guide voiced by IVONA Ian Sharpe, developer of The Talking TV Guide, trained as an IT Systems

Architect with a London-based, global IT services company, honing his skills on multimillion dollar projects ranging from Facial Image Recognition to telecommunications to data capture for the insurance industry. Ian is also a member of the Great Britain Paralympic Team since 1988, initially as a swimmer and more recently as a

cyclist. He holds 10 Paralympic medals and retains the British able-bodied record for the 4 km tandem pursuit. Ian is currently based in his native Isle of Man, applying his varied expertise to solving web accessibility issues. Visually impaired users the world over look forward to his next contribution.

LinguaSys new translation applications for IBM social collaboration applications: LinguaSys announced a new suite of language translation applications, TGSocial, TGNotes, and

TGChat, for IBM Lotus and LotusLive social collaboration applications, both on-premises and in the cloud (IBM Collaboration Solutions (www-01.ibm.com/software/info/itsolutions/collaboration). TG Social provides users with real time translation for individually highlighted threads in IBM Lotus Connections. TGNotes enables IBM Lotus Notes e-mails to be translated on the fly. TGChat allows IBM Lotus Sametime participants to chat using their native language.

Page 33: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 33

Accentus snaps up speech recognition firm Mrecord: Mrecord was acquired by Canadian company Accentus, which plans to bolster its own software offerings with Mrecord's speech recognition and transcription technology . Raleigh, North Carolina-based Mrecord‘s medical records technology includes software that can transcribe speech into text. Accentus said that Mrecord‘s technology will enhance the Ottawa-based company‘s ambulatory care service offering. Financial terms of the deal between the two privately held companies were not disclosed. The Mrecord acquisition is the third in recent months for Accentus. In December, the company completed acquisitions of medical transcription companies Zylomed Transcription and Transolutions. The deals broadened Accentus‘ technology offerings as well as its customer base in the United States; Zylomed serves hospitals, clinics and physicians‘ offices in more than 35 states. WebChartMD Partners with M*Modal to Enhance Transcription Workflow Platform with Speech Understanding™ Technology

M*Modal’s Speech Understanding Brings Powerful Productivity Capabilities to WebChartMD Customers: M*Modal and WebChartMD announced the completion of M*Modal‘s Speech Understanding technology integration into the WebChartMD platform. M*Modal‘s speech understanding technology, now part of WebChartMD‘s robust transcription workflow platform, is already being adopted by the almost 100 medical transcription service organizations currently using the WebChartMD platform. The announcement is made in conjunction with the Clinical Documentation Industry Association (CDIA) Annual Conference in Charlotte, NC. As an advanced application service provider that works with hospitals, clinics and transcription companies, WebChartMD utilizes M*Modal‘s Speech Understanding platform to provide an integrated back-end speech solution for its clients who need to drive efficiencies by turning transcriptionists into highly productive editors. This move reflects WebChartMD‘s strategy of incorporating advanced technologies into its end-to-end workflow system, and the selection of M*Modal brings a proven platform relied on extensively by most of the leading MTSO‘s in the industry. M*Modal works with a wide variety of clinical documentation, RIS/PACS, EHR, and other software and service partners who leverage the company‘s unique Speech Understanding platform.

Nuance Healthcare Development Platform will bring medical speech recognition to the cloud in 23 languages: Nuance announced that the Nuance Healthcare Development Platform, which was debuted at HIMSS11 in Orlando, Florida, will be made available for use in International markets. With this new cloud-based platform, third-

party developers can rapidly incorporate Nuance Healthcare technologies as part of a wide variety of healthcare applications, providing users with new capabilities for speech-driven clinical documentation and voice navigation. The Nuance Healthcare Development Platform for use in International markets will be introduced by the end of the year, and will support 23 recognition languages. The Nuance Healthcare Development Platform supports multiple devices and operating systems. By providing all of the CPU intensive speech processing in the cloud, this new platform enables Healthcare ISVs (Independent Software Vendors) and internal development teams at provider organisations to seamlessly embed the capabilities of Nuance SpeechAnywhere, powered by SpeechMagic, as part of browser-based applications, mobile applications and thin-client desktop solutions. The Nuance Healthcare Development Platform will extend the current Nuance speech recognition offering and will be available to existing SpeechMagic as well as Dragon Medical partners and users as an add-on option. While at a patient‘s bedside, in their office or at home, doctors can simply dictate a search query via a securely encrypted channel to access desired information. Whether the clinician is looking for updated patient test results or clinical information from an EHR, clinical data repository or other source of medical content, it can be easily accessed via a mobile device.

Page 34: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 34

Patients too can benefit from cloud-based speech recognition, for instance by logging on to hospital web portals and speak their key medical information, including history of present illness, current medications and allergies. Nuance Healthcare will present previews of the Nuance Healthcare Development Platform with Nuance SpeechAnywhere at these key industry events: Spain: Congreso Nacional de Hospitales (Madrid, 26-29 April), booth 39/40 Hungary (pan-European): eHealth Week (Budapest, 10-12 May), booth 521 France: HIT (Paris, 17-19 May), booth A71 The first pre-release of the Nuance Healthcare Development Platform for International markets will be made available to Nuance partners at the Nuance Healthcare Partner Event on October 6 and 7, 2011. Nuance integration partners, whether using Dragon Medical or SpeechMagic, will then have the option to expand their portfolio by offering the Nuance Healthcare Development Platform on top of their current solution.

Nuance’s PowerScribe Achieves DIACAP Certification: Nuance Communications' radiology speech recognition reporting platform, PowerScribe, has been certified for compliance with U.S. Department of Defense (DoD) security requirements, as tested and verified by the Department of Defense Information Assurance Certification and

Accreditation Process (DIACAP). PowerScribe, which is used by more than one-third of all radiologists in the United States, is now distinguished as a DoD-approved healthcare IT solution; it will be used across Air Force, Army and Navy Networks. DIACAP certification is an intensive and stringent requirement for all IT systems that operate on a DoD network.

Philips Voice Tracer 600: Royal Philips Electronics announced the new Voice Tracer 600 series recorder. The memory capacity of all new Voice Tracers has been doubled to up to 4GB, which allows for 44 days of constant voice recording. The Voice Tracer 625 now includes the new Nuance Dragon Naturally Speaking 11 speech recognition software to convert the recorded speech to text.

Scribe Succeeds in Delivering Speech to the Ambulatory Transcription Market: Scribe Healthcare Technologies, Inc., trumpets its success utilizing Scribe's unique implementation of M*Modal Speech

Understanding engine on the Scribe medical documentation platform. Unlike other platforms offering the M*Modal solution, Scribe redesigned its platform to fully integrate all aspects of the M*Modal technology while creating complementary tools that result in the first fully SaaS offering targeted at both In-Patient and Ambulatory markets. Scribe ensures success by making the process fully accountable and using the data to improve transcriptionist productivity. As organizations look to technology to reduce documentation costs, it is also important to keep an eye on extracting greater value from the transcribed data. The Scribe implementation of M*Modal offers cost reduction and document formatting in an HL7 Clinical Documentation Architecture. This dual benefit ensures compatibility with the EMR/EHR.

SpeechGear expands translation Capabilities: SpeechGear announces the immediate availability of their new remote server module that extends SpeechGear‘s patented instant translation capabilities to 35 languages. Versions of SpeechGear‘s entire Compadre® suite of conversation and document translation products are now available that enable instant bi-directional translation between American or British English and the following languages: • Arabic • Bulgarian • Chinese Simplified • Chinese Traditional • Czech • Danish • Dutch • Estonian • Finnish • French • German • Greek • Hebrew • Hungarian • Indonesian • Italian • Japanese

Page 35: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 35

• Korean • Latvian • Lithuanian • Norwegian • Polish • Portuguese • Russian • Slovak • Slovenian • Spanish • Romanian • Swedish • Thai • Turkish • Ukrainian • Vietnamese With these products, you can now hold conversations with individuals using any of these languages, plus instantly translate documents such as Microsoft Word files or PowerPoint presentations. To support individuals with disabilities, the remote server module will also soon be available with the Interact-AS Suite that has been developed by SpeechGear‘s subsidiary, Auditory Sciences.

Highlands and Islands go digital with letters: NHS Highlands and Islands sites won‘t just be benefiting from dictation management software WinScribe recently installed by Voice Technologies – they will also see major savings from the company‘s WinVoicePro software in 2011.

The dictation solution that speeds up the generation of documents will save the health board an estimated 63p every time a letter is created and sent electronically through the system.

Industrial Speech Applications/Products/Services/Partnerships/Financials

Aldata Demonstrates Voice Directed Warehousing: Aldata Solution demonstrated its Voice Directed Warehousing solution, used to improve productivity, accuracy and traceability with multi-modal voice technology. Whether replacing or using an existing warehouse solution, Aldata‘s Voice Directed

Warehousing easily plugs in and serves as an important means for capturing vital data in order to ensure compliance with the latest industry and government traceability regulations.

Datalogic Mobile unveils rugged mobile computer: Datalogic Mobile debuted the Falcon X3, a rugged data terminal that's available in both handheld and pistol-grip versions. The company introduced the new mobile computer at the Material Handling Industry of America's 2011 ProMat Show. The Falcon X3 comes with either a Windows CE or Windows Mobile 6.5 operating system. The unit supports wifi or Bluetooth Class-II wireless communications. The terminal comes with optional 3-megapixel autofocus camera on the Windows Mobile model. All units come preloaded with Wavelink Avalanche for device management as well as Terminal Emulation software that allows the device to be used with an older warehouse management system (WMS) application.

Roche, Merrill Corporation Implement Datria Voice Solutions: Datria Systems announced Roche and Merrill Corporation

have integrated the company's voice-enabling technology to increase productivity and create efficiencies in their supply chains. Roche, a leader in research-focused healthcare with combined strengths in pharmaceuticals and diagnostics, has implemented Datria Voice in its Indianapolis facility to voice enable existing handheld devices for container and cart picking for its diabetes care and diagnostic products in a variety of warehouse zones. Freeing the users' hands allows them to focus on the job versus the tool, and has resulted in increased pick rates while maintaining the same level of accuracy for Roche. Merrill Corporation provides the world's leading legal, financial and corporate professionals with a suite of advanced services as well as secure web-based tools to gather, organize and manage the confidential and time-sensitive information related to their matters or transactions. The company is using Datria Voice for filling document requests which enables Merrill to batch pick fulfillment requests, improve efficiency and increase order fulfillment rates for its customers.

Datria offers device-agnostic voice solution: Datria Systems Inc. displayed its voice software for warehouse, field service, and manufacturing operations

at ProMat 2011 today. The software, which runs on a server, can communicate with workers through standard devices such as cell phones or handheld terminals. Because no special unit is required,

Page 36: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 36

Datria Vice President Brad Wyland says that its voice-technology approach is cheaper than other solutions.

Dematic and Crown Equipment Alliance Launches LaserTrucks+ Solution: Dematic and Crown Equipment Corp., a forklift manufacturer, are set to collaborate on the research and development of new products that improve

warehouse and distribution center productivity. The two companies have introduced the first product resulting from the alliance, LaserTrucks+, an integrated solution that combines the Dematic picking and voice software suite and Automated Guided Vehicle (AGV) guidance technology, applied to a Crown PC 4500 Series rider pallet truck that will increase case picking productivity in the warehouse.

RedPrairie And Lucas Systems Team Up For Voice Directed Warehouse Management: Lucas Systems announced that it has entered into a reseller agreement with RedPrairie. Under the agreement, RedPrairie will resell Jennifer VoicePlus applications that help maximize the productivity and accuracy of picking and other warehouse activities working in concert with RedPrairie Warehouse Management solutions. Jennifer is already installed with a RedPrairie WMS at OM HealthCare Logistics, an Owens & Minor business unit and an innovative healthcare third-party logistics (3PL) provider. Jennifer VoicePlus is the first voice application to use voice and scanning interchangeably in a single application, to interleave voice-picking with other tasks, and to run on standard mobile computing devices that can be used in non-voice processes. Under this reseller agreement, Lucas Systems and RedPrairie have designed a standard interface that allows Jennifer to interact directly with the RedPrairie WMS, allowing companies to rapidly deploy voice-optimized WMS applications. Jennifer, by Lucas, is the leading hardware-independent voice solution in North America, with tens of thousands of users at hundreds of DCs.

LXE rolls out vehicle-mount computer: LXE announced the launch of a new vehicle-mount computer that offers multiple logistics applications. LXE, which made the announcement at ProMat 2011 in Chicago, said the computer, called the "Thor, " is the latest-generation vehicle-mount computer to use the half-screen, full keypad form factor that Norcross, Ga.-based LXE pioneered more than 30 years ago. One of The Thor computer's most distinguishing features is field-replaceable front panel. With ordinary vehicle-mount computers, any significant damage to front panels would require end users to return the entire computer to the manufacturer for front-panel replacement, or to purchase an entirely new unit. In comparison, users can replace Thor's front panel themselves with simply a screwdriver. As a result, users save both time and money.

Psion debuts new handheld: Psion has released the Omnii XT10 handheld computer with scanner. It runs on the Windows CE operating systems and uses 802.11 b/g for wireless communications. The device offers an interchangeable keyboard. Depending on the configuration (scanner, radio, keyboard), the unit ranges from $2,500 to $3,300.

RedPrairie to include Vocollect VoiceArtisan: RedPrairie Corp. said it will add VoiceArtisan technology from Vocollect as a key enhancement to its Warehouse Management 2011.1 release. VoiceArtisan enables faster development of voice

workflows and reinforces RedPrairie's efforts in the area of voice workflow development. By utilizing VoiceArtisan, RedPrairie will be able to expand its voice-enabled selection, put-away, replenishment, loading, and cycle-counting functionality.

topVOX's BlueMaster Headset Works with its topSpeech-Lydia Voice Picking Solution: topVOX has said that the new BlueMaster headset works with its topSpeech-Lydia voice picking solution.

Page 37: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 37

The company said that this combination improves ergonomics and safety by eliminating headset wires with Bluetooth wireless technology. The design of a topSPEECH-Lydia is a sophisticated speech-controlled picking system which outperforms any common MDE or paper-based system by ―significantly‖ increasing both efficiency and quality, said company officials. It features both a speaker-dependent as well as a speaker-independent recognition, thus allowing an optimized and reliable speech control in all working environments and conditions. topVOX also made sure the headset -- in combination with the Noisemaster feature of the voice solution -- eliminates background noise so that pickers can speak at a normal voice level when confirming picks.

Intermec, Vocollect plan integrated offerings: Earlier this month, Intermec Inc. completed is acquisition of Vocollect. Company executives at the ProMat 2011 show in Chicago spoke about how they expect the alliance will help both

companies expand globally. Intermec, a manufacturer of scanning and other data capture tools, announced the acquisition in January. At the time, Intermec President Patrick J. Byrne said joining the companies together would strengthen Intermec‘s position as a provider of rugged mobile tools. Thomas Murray, vice president of product management and marketing for Vocollect—which will operate as a business unit of Intermec but continue to be based in Pittsburgh—said the combined companies will work with customers to develop ―technology solutions that make the most sense for them.‖ He added that Vocollect will also continue to work with other strategic data capture companies such as Psion Teklogix. In addition, Eusterman said, the combined companies will focus increasing attention on the development of software and services. Murray said in conversations with customers, the common themes he hears are requests to reduce operating costs and support growth. Eusterman added that during the downturn, customers focused on cost and productivity issues. ―Now they are focusing on customer satisfaction.‖

Portable voice from Voxware: Voxware is promoting portability for voice recognition

solutions. ―We believe there is a difference between supporting various device platforms from hardware vendors like LXE and Motorola and a truly portable platform,‖ said Stephen Gerrard, VP of marketing and strategic planning. ―With a truly portable voice solution, you can work on a Motorola device today and switch to a new device from LXE tomorrow, all without rewriting a line of code.‖ Gerrard adds that a current customer put Voxware‘s solution to the test by loading it onto devices from three manufacturers before choosing a hardware vendor. ―They wanted to be certain they could move the solution if their needs change in the future,‖ Developments like portability of software represent a maturing voice recognition industry. ―In the past, it was OK to implement a proprietary voice system because there weren‘t the options that there are today in the market,‖ Gerrard said. ―Today, IT managers want to be able to run the solution on whatever hardware the want to run it on. Portability makes that possible.‖

Wavelink Demonstrates Speakeasy adds: Wavelink Corporation announced it will be demonstrating several enhancements to its Speakeasy voice solution at ProMat 2011 in Chicago. The new features in the Speakeasy platform provide greater flexibility and further accuracy to voice recognition while increasing the speed of implementation. Speakeasy remains a speaker-independent solution in which any user can pick up any mobile device with the Speakeasy client and utilize the voice features. Now, companies are able to have individual users train the speech-to-text engine to their particular voice and speech patterns. The voice training is saved and managed via Wavelink Avalanche as a voice profile, so that any user can log in to any device and have it equipped with their voice training profile. In certain environments this option improves the accuracy of the voice recognition and still maintains the user flexibility to operate any mobile device. The following new features of Speakeasy also help improve productivity and accuracy: User calibration of the speech recognition engine for a particular hardware and environment

Page 38: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 38

Real-time speech confidence level viewer lets users diagnose the cause of inaccuracies in recognition Speech status indicator ensures users are speaking when the speech engine is listening Screen Reformatter enables administrators to designate specific fields for speech-to-text and text-to-speech entry User interface can be tailored to specific users or applications

Merck improves multiple warehouse operations with just 8 voice pickers

To optimise time consuming warehouse processes and ensure compliance with system safety requirements, Merck has implemented Zetes' 3i Voice solution (3iV). After voice enabling just 8 operators, important business improvements have been seen including increased productivity, improved operator satisfaction, reduced error rate and improved planning. Merck's original warehouse processes based on barcode scanning were time consuming and the equipment used no longer complied with the company's systems safety requirements. In 2010, they implemented a 3i Voice solution to improve process efficiency. "We had already worked with Zetes in the past, and we were aware that they implemented voice picking systems. During a site visit to an existing installation, we saw first hand the advantages of adopting this technology", explains Luis Azevedo, Merck's IT Manager. Zetes' impressive track record of successful voice implementations both in Portugal and internationally played an important role in Merck's decision to choose the company as its partner.

Base Speech Technology Applications/Products/Services/Partnerships/Financials

APH News, April 2011: The monthly newsletter from the American Printing House for the Blind (APH) is on-line.

AT&T at CSUN: Innovative communication technology designed to address the needs of aging adults and persons with disabilities can also benefit consumers without physical, cognitive and other limitations, AT&T reported at the 26th Annual International Technology & Persons with Disabilities Conference, March 14-19, in San Diego, hosted by the Center on Disabilities at California State University, Northridge (CSUN). At the conference, AT&T highlighted several emerging technologies, ranging from ―smart shoes‖ communications that monitor for compromises in a patient‘s balance, to devices with touchscreens that can be used without the benefit of sight. AT&T will also introduce a new accessible web video highlighting the array of products and services it makes available to the disability and aging adult communities. The video is available at www.att.com/accessibility. The annual CSUN conference comes on the heels of an extensive study on AT&T‘s commitment to the disability and aging communities. The Global Initiative for Inclusive ICTs (G3ict), a flagship advocacy initiative of the United Nations Global Alliance for ICT and Development, recently selected AT&T and issued the report highlighting AT&T‘s efforts to integrate Universal Design (designing products and services for the greatest number of users) and accessibility in product development, customer services and other critical business areas. The study, ―Accessibility, Innovation and Sustainability at AT&T,‖ documents the company‘s efforts to create a more inclusive workplace, particularly one that promotes innovation, sustainability, and independence for customers living with disabilities. The G3ict study describes how AT&T has effectively integrated the principles of Universal Design and accessibility into its business, specifically around product development, human resources, talent attraction, recruitment, marketing and customer service. Furthermore, the case study highlighted AT&T‘s approach to accessibility as an important element to the company‘s citizenship and sustainability strategy.

Nuance shares rise on Apple iPhone report: Nuance shares closed up 7 percent on Wednesday after an analyst said the company is expected to be Apple Inc.'s Shares rose to $19.55 on Wednesday following the report from Soleil Securities. Shares in the company were flat as of the market close on Thursday.

Page 39: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 39

For Apple's IOS 5.0 for the next-generation iPhone, Soleil wrote that it believes Apple's partner for the phone's voice recognition technology is Nuance. "We believe that the depth of the Apple/Nuance relationship is far deeper than Wall Street realizes and will serve as a significant catalyst for the shares," the firm wrote.

Table P1

March, 2011 Speech Technology Patents Awarded by US PTO

Assignee USPTO # Title

Sony Corp. 7,917,930 Information signal transmission system and remote control device for the same

Symbol Technologies 7,917,888 System and method for building multi-modal and multi-channel applications

Audible Magic Corp 7,917,645 Method and apparatus for identifying media content presented on a media playing device

Verizon 7,917,581 Call completion via instant communications client

Microsoft 7,917,514 Visual and multi-dimensional search

Limelight Networks 7,917,492 Method and subsystem for information acquisition and aggregation to facilitate ontology and language-model generation within a content-search-service system

Seven Networks 7,917,468 Linking of personal information management data

Mitsubishi Electric 7,917,368 Method for interacting with users of speech recognition systems

VoiceBox Technologies

7,917,367 Systems and methods for responding to natural language speech utterance

Nuance Comm's 7,917,365 Synchronizing visual and speech events in a multimodal application

Hewlett-Packard 7,917,364 System and method using multiple automated speech recognition engines

Telstra Corp. 7,917,363 System for predicting speech recognition accuracy and development for a dialog system

Oki Electric 7,917,359 Noise suppressor for removing irregular noise

Microsoft 7,917,357 Real-time detection and preservation of speech onset in a signal

Google 7,917,355 Word detection

IBM 7,917,350 Word boundary probability estimating, probabilistic language model building, kana-kanji converting, and unknown word model building

Thomson Licensing 7,917,336 Geometric source separation signal processing technique

Sony Ericsson 7,917,178 Wireless communications device with voice-to-text conversion

Open Invention Network

7,916,852 Method and system for providing quick directions

Securus Technologies 7,916,845 Unauthorized call activity detection and prevention systems and methods for a Voice over Internet Protocol environment

Verizon 7,916,842 Systems and methods for providing voicemail services including caller identification

Treyz; G. Victor et al 7,916,579 Digital audio devices

Applied Minds 7,916,165 Systems and method for enhancing teleconferencing collaboration

United Video Properties

7,913,278 Interactive television program guide with remote access

IBM 7,913,155 Synchronizing method and system

Verint 7,913,063 System and method for performance based call distribution

Research In Motion 7,912,984 System and method for generating a wireless application from a web service definition

AT&T 7,912,963 Methods and apparatus to control a voice extensible markup language (VXML) session

Page 40: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 40

IBM 7,912,727 Apparatus and method for integrated phrase-based and free-form speech-to-speech translation

AT&T 7,912,726 Method and apparatus for creation and user-customization of speech-enabled services

Adobe Systems 7,912,724 Audio comparison using phoneme matching

Nuance Comm's 7,912,721 System and method for automatic speech recognition

AT&T 7,912,720 System and method for building emotional machines

AT&T 7,912,718 Method and system for enhancing a speech database

Galick; Albert 7,912,717 Method for uncovering hidden Markov models

Sony Corp. 7,912,716 Generating words and names using N-grams of phonemes

Nokia Corp. 7,912,715 Determining distortion measures in a pattern recognition process

Nuance Comm's 7,912,714 Method for segmenting communication transcripts using unsupervised and semi-supervised techniques

Loquendo 7,912,713 Automatic speech recognition system and method using weighted confidence measure

Philips Electronics 7,912,708 Method for controlling duration in speech synthesis

Microsoft 7,912,707 Adapting a language model to accommodate inputs not found in a directory assistance listing

IBM 7,912,703 Unsupervised stemming schema learning and lexicon acquisition from corpora

Phoenix Solutions 7,912,702 Statistical language model trained with semantic variants

Microsoft 7,912,700 Context based word prediction

AT&T 7,912,699 System and method of lattice-based search for spoken utterance retrieval

Sony Corp. 7,912,696 Natural language processing apparatus and natural language processing method

Donnelly Corp. 7,912,646 Driver assistance system for vehicle

Intelligent Technologies

7,912,645 Information transfer arrangement and method for vehicles

AT&T 7,912,642 Automated location-intelligent traffic notification service systems and methods

Navteq 7,912,638 Method of collecting information for a geographic database for use with a navigation system

Microsoft 7,912,637 Landmark-based routing

Denso Corp. 7,912,632 Navigation system

Garmin Int'l 7,912,592 Automatic speech recognition system and method for aircraft

AudioCodes 7,912,567 Noise suppressor

Samsung 7,912,519 Mobile communication device with function-assignable side key and method for controlling the side key

Intellisist 7,912,512 Sharing account information and a phone number between personal mobile phone and an in-vehicle embedded phone

Microsoft 7,912,503 Smart interface system for mobile communications devices

Jumptap 7,912,458 Interaction analysis and prioritization of mobile content

Reqall 7,912,454 Method and system for archiving data in real-time communications

AT&T 7,912,447 Home gateway system for home automation and security

Master; Aaron 7,912,232 Method and apparatus for removing or isolating voice or instruments on stereo recordings

Volkswagen 7,912,228 Device and method for operating voice-supported systems in motor vehicles

Avaya 7,912,207 Data messaging during telephony calls

AT&T 7,912,201 Directory assistance dialog with configuration switches to switch from automated speech recognition to operator-assisted dialog

AT&T 7,912,187 Transcoding voice to/from text based on location of a communication device

Page 41: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 41

Microsoft 7,912,186 Selectable state machine user interface system

Earthcomber 7,911,335 System and method for locating and notifying a mobile user of people having attributes or interests matching a stated preference

Ellis; Michael et al 7,909,737 Workout definition and tracking methods

Comcast 7,908,628 Video and digital multimedia aggregator content coding and formatting

Katsuranis; Ronald M. 7,908,565

Voice activated system and method to enable a computer user working in a first graphical application window to display and control on-screen help, internet, and other information content in a second graphical application window

AOL 7,908,554 Modifying avatar behavior based on user action or mood

TellMe Networks 7,908,383 Method and apparatus for phone application state management mechanism

AT&T 7,908,381 Simultaneous visual and telephonic access to interactive information delivery

Health Hero Network 7,908,152 Blood glucose monitoring system

Funai Electric 7,908,146 Digital television receiver controlled by speech recognition

Engate 7,908,145 Down-line transcription system using automatic tracking and revenue collection

IBM 7,908,143 Dialog call-flow optimization

Sony Corp. 7,908,142 Apparatus and method for identifying prosody and apparatus and method for recognizing speech

IBM 7,908,141 Extracting and utilizing metadata to improve accuracy in speech to text conversions

Samsung Electronics 7,908,139 Apparatus and method of reducing noise by controlling signal to noise ratio-dependent suppression rate

Sony Corp. 7,908,137 Signal processing device, signal processing method, and program

Microsoft 7,908,132 Writing assistance using machine translation techniques

Leupold & Stevens 7,908,083 System and method for recording a note with location information derived from rangefinding and/or observer position

Jumptap 7,907,940 Presentation of sponsored content based on mobile transaction event

Intuit 7,907,705 Speech to text for assisted form completion

Avaya 7,907,704 Caching user information in an integrated communication system

Intellectual Ventures 7,907,703 Method and apparatus for telephonically accessing and navigating the internet

Inter-Tel 7,907,702 Communication device for visually impaired persons

Ellis; Michael et al 7,905,815 Personal data collection systems and methods

Harrison, Jr.; Shelton E.

7,905,769 Integrated game system, method, and device

Intel 7,904,879 Reorganized storing of applications to improve execution

Sharp 7,904,814 System for presenting audio-video content

Microsoft 7,904,797 Rapid media group annotation

Accenture 7,904,501 Community of multimedia agents

VideoMining 7,904,477 Object verification enabled network (OVEN)

Kana Software 7,904,414 Adaptive multi-channel answering service for knowledge management systems

IBM 7,904,399 Method and apparatus for determining decision points for streaming conversational data

Denholm; Diana B. 7,904,312 Patient communication method and system

Health Hero Network 7,904,310 Blood glucose monitoring system

Nuance Comm's 7,904,300 Supporting multiple speech enabled user interface consoles within a motor vehicle

Nuance Comm's 7,904,299 Method, system, and apparatus for monitoring security events using speech recognition

Page 42: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 42

Rao; Ashwin P. 7,904,298 Predictive speech-to-text input

Robert Bosch GmbH 7,904,297 Dialogue management using scripts and combined confidence scores

Nexidia 7,904,296 Spoken word spotting queries

Coelho; Rosangelo Fernandes

7,904,295 Method for automatic speaker recognition with hurst parameter based features and method for speaker classification based on fractional brownian motion classifiers

AT&T 7,904,294 System and method for processing speech recognition

Toshiba 7,904,291 Communication support apparatus and computer program product for supporting communication by performing translation between languages

HTIIP, LLC 7,904,219 Peripheral access devices and sensors for use with vehicle telematics devices and systems

Hoffberg; Steven 7,904,187 Internet appliance system and method

InterDigital Technology 7,904,113 Method and apparatus for detecting and selectively utilizing peripheral devices

IBM 7,903,800 Method and system for visually representing telephone call tree interface

Applied Voice & Speech Technologies

7,903,793 Template-based electronic message generation using sound input

Enterprise Integration Group

7,903,792 Method and system for interjecting comments to improve information presentation in spoken user interfaces

American Calcar 7,902,969 Technique for operating a vehicle effectively and safely

Health Hero Network 7,901,625 System for performing diabetes self-care

BrainGate Co. 7,901,368 Neurally controlled patient ambulation system

Sony Corp. 7,900,256 Communication apparatus, communication method, and recording medium used therewith

IBM 7,900,133 Annotation structure type determination

Reisman; Richard 7,899,915 Method and apparatus for browsing using multiple coordinated device sets

Trandal; David S. 7,899,823 Methods and systems for inventory management

Clairvista LLC 7,899,774 System and method for interactively delivering expert information to remote outlets

Intel 7,899,771 Most probable explanation generation for a Bayesian Network

Accenture 7,899,698 Real-time sales support and learning tool

The US Army 7,899,687 System and method for handling medical information

Bevocal LLC 7,899,675 System, method and computer program product for transferring unregistered callers to a registration process

The US Navy 7,899,674 GUI for the semantic normalization of natural language

Microsoft 7,899,673 Automatic pruning of grammars in a multi-application speech recognition interface

Nuance Comm's 7,899,672 Method and system for generating synthesized speech based on human recording

Avaya 7,899,671 Recognition results postprocessor for use in voice recognition systems

eScription 7,899,670 Server-based speech recognition

Gadbois; Gregory John

7,899,669 Multi-voice speech recognition

Enounce 7,899,668 Method and apparatus to prepare listener-interest-filtered works

Securus Technologies 7,899,167 Centralized call processing

Speech Industry People

Albert Kim has been tapped to lead the Mobile Division of SpeechCycle and will be responsible for spearheading this new business unit.

Page 43: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 43

Financials

Table F1

Q4 Quarterly Results

Supplier Item

Q4, 2010

2009 to 2010 % change

Q3, 2010 to Q4, 2010 %

change

Q4, 2009

Q3, 2010

$M $M

Call Genie

Total Revenue 1.10 37.50% -15.38% 0.80 1.30

Net Income -1.00 -62.96% 0.00% -2.70 -1.00

Cash or equiv. 0.40 -88.57% -20.00% 3.50 0.50

Verint

Total Revenue 186.869 8.05% 0.12% 172.74 186.641

Net Income 9.511 236.78% -68.71% -22.22 30.393

Cash or equiv. 169.906 -25.20% 23.23% 184.34 137.879

Wizzard Software

Total Revenue 1.45 -1.47% -1.60% 1.47 1.47

Net Income -0.84 -13.58% 8.86% -0.97 -0.77

Cash or equiv. 0.27 -64.74% -55.48% 0.76 0.60

Table F2

Market Price Summary April 1, 2011

Company Symbol

Share price 2010 Market Cap. Mkt

Cap./ Rev.

2009

Last 52-wk. Rev. Net

Income 4/1/2011

% change

Rev.

Hi Low $M $M

Comverse Technology

CMVT 7.14 9.3 4.59 NA NA 1,450.00 -3.33 NA NA

Convergys CVG 14.49 15 9.5 2,203.4 -53.2 1,777.00 4.53 0.81 2,421.00

DynaVox DVOX 5.48 19.2 3.8 91.16 8.838 51.41 -3.87 81.44

Enghouse (Canada)

ESL 10.54 10.4 7.6 94.2 10.2 265.31 17.06 2.82 78.4

Call Genie GNE 0.17 0.27 0.05 4.2 -8.5 12.85 -14.73 3.06 3.90

Interactive Intelligence

ININ 39.4 40.39 13.77 166.315 14.901 725.06 21.07 4.36 131.418

Medquist MEDQ 10.41 12.39 7.0001 NA NA 390.95 15.92 NA NA

Nice Systems (Israel)

NICE 37.25 37.5 25.1 689.451 48.707 2,300.00 10.90 3.34 583.115

Nuance Comm's NUAN 19.56 20.97 14.45 1118.9 -19.1 5,870.00 9.51 5.25 950.4

Soundbite Comm's SDBT 2.65 3.38 2.4 39.494 -3.285 43.42 8.20 1.10 40.183

Transcend Services

TRCR 23.9 24.42 12 94.307 8.52 253.05 14.43 2.68 71.764

Verint VRNT 36.3 38.1 19.63 726.799 11.403 1,330.00 6.40 1.83 703.633

Wizzard Software WZE 0.245 0.44 0.1601 5.54 -4.11 17.33 0.81 3.13 5.194

Total Total 5,233.8 14.37 14,486.38 8.16 2.77 5,070.45

Page 44: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 44

Phone: 949.366.9000 ▪ www.SpeechTekJobs.com ▪ Email: [email protected]

Sr. Program Director/ Principal Investigator – Speech & Language Technology Research Location: Silicon Valley (Relocation assistance is available.) Position ID: AXG-1313 Career opportunity to play a key role with an industry-leading speech and language technology research laboratory. The lab performs leading-edge research in speech recognition, automatic spoken language translation, speaker recognition and verification, human-machine interfaces, and other areas of speech/language technology. The successful candidate will take a leadership role in establishing research programs in speech and language technology, and will contribute to ongoing work with new technical approaches, designs, and applications. Will lead business development efforts, including preparation of winning proposals, collaboratively define new research directions, and lead technical project and proposal teams. Experience winning US Government proposals is highly desirable.

Staff Software Engineer – Speech Recognition Location: Silicon Valley, CA Position ID: AXG-1318

DESCRIPTION:

Opportunity for top-notch R&D engineer to join an industry-leading voice team to build a cutting edge natural language platform. Looking for a creative person who enjoys exploring new technologies and is passionate about building intelligent systems to revolutionize the way humans and machines interact. RESPONSIBILITIES:

The most important skills for this position are:

• Expert on server side system architecture design and development. Will build large scale systems in the cloud.

• Experience with the speech recognition engine and integration

• Familiarity with Natural Language Understanding (NLU), language modeling, and machine learning technologies (these are not mandatory but a plus)

Page 45: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 45

REQUIREMENTS:

• Ph.D. or Master in CS or EE with 4+ years of relevant working experience

• In-depth knowledge and experience in Statistical Language Modeling, NLU, machine learning

• Familiar with research tools like Matlab, R, etc.

• Good hands-on coding skills in Java or C++, scripting language

• Experience in speech technologies a big plus

• Excellent communication and interpersonal skills

Speech Research Engineer Location: Boston, MA (Relocation assistance is available.) Position ID: AXG-1315

US citizenship required

Company is responsible for the ongoing research and development of speech products, delivering audio processing applications to enable live or recorded speech to be manipulated, edited, searched and hyperlinked as easily as text. This is achieved with a wide range of speech processing technologies from audio segmentation, speaker and language identification through automatic speech recognition and understanding.

Position Summary

As a member of a small team of research and development engineers you will work both independently and in teams developing algorithms and models for a wide range of audio analysis problems.

Experience with speech recognition systems and algorithms, especially acoustic and/or language model development is essential. You will need strong C++ and Python programming, as well as Linux shell scripting skills.

Language/linguistic skills are also highly desirable.

In addition to the technical skills mentioned above, the successful candidate will have a proven ability to work with minimal supervision as part of a team located in both Boston MA and Cambridge UK, a proactive personality, and excellent oral and written communication skills. Occasional travel may be required.

Candidates must be a US Citizen.

Speech Researcher / Acoustic Modeling Location: Boulder, CO Position ID: AXG-1316 This Speech R&D Lab conducts research in speech recognition and processing technologies and is offering a unique career opportunity to work with world-class experts on meeting their goals in

Page 46: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 46

advanced speech research and development. As a member of Lab, you will be responsible for inventing new technologies that will enable and drive future products, including developing advanced algorithms for speech recognition, acoustic modeling, speech analysis and synthesis, as well as assisting in development and prototyping human language technologies for technology-enabled language learning. DUTIES & RESPONSIBILITIES:

• Work collaboratively with world-class speech recognition experts to define and meet speech research and development goals.

• Assist in the company’s speech processing development by implementing and developing new cutting-edge algorithms for speech recognition, speech analysis & speech synthesis.

• Assist in developing and prototyping human language technologies for effective pronunciation feedback.

IVR Developer / VXML, C#, XSL, Javascript Programming Location: Pasadena, CA area Position ID: AXG-1271 Job Description: The IVR Developer is responsible for the detailed design, construction, implementation, and support of telephony Speech- and DTMF-based user interface systems that meet established specifications, design criteria, quality standards and business requirements. With global Call Centers in three continents, the IVR is the callers' first point of contact, and therefore, is a mission-critical application to the organization.

• The developer will have primary responsibility for coding in VXML, C#, Javascript, and XSL as needed to implement various callflows.

• Will work closely with the Business Systems Analyst (who generates and updates business

requirements and create all related callflows) to assure the programming conforms to each requirements doc and callflow diagram.

• Will write test procedures and perform unit testing; integrate with other system components;

perform debugging, troubleshooting, QA and production support.

• Will design some of the upfront voice user interface details in support of the Business Systems

Analyst, especially during initial vetting of requirements with the internal company stakeholders.

• Responsible for coding new projects with new designs as well as maintaining/updating existing

applications that do not involve much new design work.

• Reports to the Manager of IVR Development

• Experience in speech technologies a big plus

• Excellent communication and interpersonal skills

Page 47: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 47

Contact Solutions is looking for quality people:

Contact Solutions, Inc. is a rapidly growing telecommunications services provider specializing in automation of customer service contacts. The company uses state of the art voice recognition and call control software to provide end to end hosted call automation services to fortune

500 clients. The Channel Sales Manager, Strategic Partners is responsible for interfacing with a key current

Contact Solutions partner to promote the sales of CSI‘s shared call automation services through the development of new joint sales opportunities and managing relationships with key partner staff in their assigned territory. The position works closely with other members of the Partner Sales and Client Services teams to successfully identify and contract new customer opportunities and promote Contact Solutions‘ offerings and ensure the highest quality customer service is delivered to the Partner and their customers.

Contact Solutions is looking for top notch Hosted Solutions sales professionals to join our growing team The Senior Account Executive is responsible for successfully selling Contact Solutions IVR services in a specific vertical industry and/or defined geographic territory. The position reports directly to the Director, Sales.

The Integration Engineer I is responsible for supporting all aspects of voice and data telecommunications as it pertains to the delivery of Contact Solutions‘ Interactive Voice Response (IVR) applications. The position reports directly to the Manager, Engineering and Operations.

The Continuous Improvement Project Manager is responsible for managing the identification, implementation, and sign-off on Continuous Improvement recommendations through the 4 phases of our methodology. Further, the PM will be responsible for managing and delivering strategic, macro-improvement projects following our SDLC process. The PM will be required to interface with internal and external clients, summarizing in written and oral deliverables the findings of their analysis. The position reports to the Director, Continuous Improvement.

The Continuous Improvement Analyst is responsible for performing detailed analyses, assessing for Best Practice application and identifying functional enhancements that improve the customer experience and/or generate cost savings opportunities. The analyst will also be responsible for the creation, tracking, and communication of these analyses, recommendations, and cost savings figures as they relate to recommended client changes. The position reports to the Manager, Continuous Improvement.

The Solution Architect is responsible for providing pre-sale discovery and solution assessments in support Contact Solution‘s field sales efforts. Primary responsibilities include leading on-site discovery sessions, developing solution recommendations and proposals, and conducting solution presentations/demonstrations.

We are seeking Project Managers responsible for driving all aspects of the delivery of Contact Solutions? Interactive Voice Response (IVR) applications.

The Contracts Manager is primarily responsible for the evaluation, development and negotiation of contracts, coordinating with internal Contact Solutions resources to ensure accuracy, and compliance with company policies and procedures. The work location will be the Contact Solutions offices‘ in Reston, Virginia. If relocation is required, associated costs will not be reimbursed by the Company.

The Staff Accountant is responsible for providing general accounting support to the department. The Human Resources Administrator I is responsible for performing day-to-day administration

of employee benefits and providing support to the company in other areas including recruitment, workforce administration, special programs and recognition. The HR Administrator I will also act as back up to the Executive Administrator in the office support function. The role reports directly to the Manager, Human Resources.

If interested, please review our current openings at http://www.contactsolutions.com/careers/ and submit your resume to [email protected].

Page 48: Automatic Speech Recognition - asrnews.com Speech Recognition March 2011 Vol. 22 No. 3 INSIDE THIS ISSUE The Emperor ... Play Antakshari on Mobile 21 Quickoffice for Android Tablet

ASRNews March 2011 48

Calendar

WERC 2010, Orlando, FL …………………………………………………………….……………………………..May 15-18, 2011 SpeechTEK Europe 2011 …………………………………………………………………….…………………….May 25-27, 2011 Middle East Call Center 2011, Dubai, ……………………………………………………………………………May 29-31, 2011 The Logistics & Supply Chain Forum, Stone Mountain, GA ………………………………………………….June 5-7, 2011 SpeechTek 2011. New York City, ………………………………………………………………………………August 8-11, 2011

Subscription Order Form

Individual Library Street _____________________________

US & Canada

Overseas US &

Canada Overseas

City/State/Zip ________________________ On-Line + Archives

$360 $360

Country _____________ Phone ________________

Hardcopy Only

$360 $410 $410 $460 To start your subscription , fill out this form and fax it to us

or send the information to [email protected] Voice Information Associates, Inc., P.O. Box 2861, Acton MA 01720, USA, Voice: 978-266-1966, Fax: 815-550-8730

Hardcopy and On-Line + Archives

$490 $550 $530 $590

___ Check enclosed. Payable to ASRNews

___ Bill me ___ Visa ___ Mastercard Credit card #: __________________________________ Name on Credit Card _____________________________ Company _______________________________________ Expires _______________________________________

Trademarks mentioned in this publication are the property of the companies mentioned; they are used editorially.

ASRNews is published monthly by Voice Information Associates, Inc., P.O. Box 2861, Acton, MA, 01720, USA, Voice. 978-266-1966,

Fax:815-550-8730, www.asrnews.com, email, [email protected] Walt Tetschner, Publisher and Editor. Subscription, annual: hard

copy only, U.S. and Canada, $360, foreign, $410, Copyright 2009, ISSN 1051-4163. All rights reserved. Reproduction or photocopying is

strictly prohibited without written permission. Voice Information Associates is an information services company specializing in the advanced

speech technology industry.