IN-DEPTH

The article explains the importance of a recent W3C Recommendation Semantic Interpretation for Speech Recognition (SISR), a new standard which completes the SRGS grammar by enabling a standard way of creating results from ASR processing.
The focus of the article is not only to describe this new language, but to outline the impact of SISR in the current world of speech applications.

>> Go to the article.

STANDARDS CORNER

• SISR is a W3C Recommendation
  April 5, 2007
  The World Wide Web Consortium released Semantic Interpretation for Speech Recognition (SISR) Version 1.0 as a W3C Recommendation. Part of a powerful trend towards Web access via interactive voice response, SISR tags are used to extract meaning from speech recognition. SISR defines the syntax and semantics of tag content in the Speech Recognition Grammar Specification (SRGS) for output as serialized XML or ECMAScript variables. Visit the Voice Browser home page.
  
  
• Voice Browser Working Group Re-chartered by W3C
  March 30, 2007
  The World Wide Web Consortium (W3C) is pleased to announce the relaunch of the Voice Browser Working Group to enable users to speak and listen to Web applications. Jim Larson (Invited Expert) and Scott McGlashan (HP) chair the group which is chartered to standardize languages for capturing and producing speech and for managing the dialog between users and computers. W3C Members may use this form to join the Working Group.
  Loquendo is an active member of the Voice Browser Working Group and fully supports its activities. Read the Voice Browser new charter for more details.
  
  
• Multimodal Interaction Working Group Relaunched Again by W3C
  March 30, 2007
  The World Wide Web Consortium (W3C) is pleased to announce the relaunch of the Multimodal Interaction Working Group to enable users to use their preferred modes of interaction with the Web. Deborah Dahl (Invited Expert) chairs the group which is chartered to develop open standards to adapt to device, user and environmental conditions, and to allow multiple modes of Web interaction including GUI, speech, vision, pen, gestures and haptic interfaces. W3C Members may use this form to join the Working Group.
  Loquendo is an active member of the Multimodal Interaction Working Group and fully supports its activities. Read the MMI new charter for more details.
  
  
• SCXML Published a New Working Draft
  February 21, 2007
  The W3C Voice Browser Working Group has published a new Working Draft of State Chart XML (SCXML): State Machine Notation for Control Abstraction.
  SCXML is a flexible state machine language that combines concepts from CCXML and Harel State Tables. It enhances the basic concept of state machines with such powerful concepts as conditions on transitions and nested and parallel states.
  SCXML can be used in many ways, including:
  • As a high-level language controlling VoiceXML 3.0's encapsulated speech modules (voice form, voice picklist, etc.)
  • As a cross-modality synchronization mechanism in a multimodal interface.
  • As a future version of the W3C call control language (CCXML).

R&D COLLABORATIONS

• Loquendo Voice Technologies for spoken Language UNderstanding in multilinguAl communication systems (LUNA project)
  To improve the effectiveness and user acceptance of automated dialogue systems in today's commercial applications based on speech recognition, new R&D activities have been devoted to the advancement in the state of the art of Spoken Language Understanding (SLU) research.
  Loquendo, partnering with University of Avignon, University of Trento, RWTH Aachen, France Telecom, CSI-Piemonte, Polish-Japanese Institute of Information Technology and Institute of Computer Science-Polish Academy of Sciences, provides its ASR technology for the 3 year EC-funded project "LUNA" (www.ist-luna.eu), addressing the problem of real-time understanding of spontaneous speech in the context of advanced telecom services.
  >> More
  
  
• COMPANIONS project: meeting in Turin and recent updates
  May 14-15, 2007
  On May 14-15, Loquendo will host the next meeting of the EU funded COMPANIONS project. The consortium, coordinated by the University of Sheffield with fifteen EU and US partners, aims at developing a virtual conversational Companion: an agent or 'presence' that stays with its user or owner for a long period (appearing to learn the owner's tastes, habits and needs) and that can provide simple services and conversations. This builds on the concept of the Embodied Conversational Agent (ECA): an animated character that communicates with users by using and understanding speech.
  One scenario of the project is the 'Senior Companion', aimed at the increasing numbers of elderly people who are in need of company, and would probably be grateful for the access to their own past that conversation with a Companion could provide. The Companion may tell jokes, (remembering what it has already told) or perhaps just chat. The Companion will also be able to organise and make sense of its owner's photographic and image 'memories' as a coherent life narrative: carrying out activities within a user's digital space on the Internet that extend the current capabilities.
  As an EC Sixth Framework Integrated Project, COMPANIONS is a federation of subprojects that brings together a set of technologies to tackle this large and complex question. Each of the project's areas (involving design and evaluation of the Companion, machine learning, personality building and emotional speech recognition and synthesis) will make separate, publishable, advances, while collaborating to produce a whole that is more than the sum of the parts. Loquendo, a partner in the project, is providing speech technologies and will be making a big contribution in research and development, mainly to deliver natural expressivity to the agent's speech.
  
  More details are available on the COMPANIONS website at: www.companions-project.org.
  

LOQUENDO Products

• Loquendo Awarded "Best Innovation in Text to Speech" at SpeechTEK West 2007
  March 8, 2007
  Loquendo announced that Loquendo TTS has been awarded "Best Innovation in Text to Speech" at AVIOS-SpeechTEK West 2007, in San Francisco, for the third year in a row. The prize was awarded for Loquendo TTS Automotive Solution, a complete and multilingual solution developed for manufacturers of GPS navigation devices and in-car telematics. The Avios-SpeechTEK 2007 award is a clear recognition of Loquendo's position at the leading edge of speech-synthesis technology.
  The AVIOS-SpeechTEK Awards are presented to companies for their innovations in technology, and are chosen by a committee appointed by board members of the American Voice Input/Output Society (AVIOS).
  >> More
  
  
• Loquendo and PIKA Technologies Certify Interoperability of Pika's Telephony Boards and Loquendo Speech Technologies
  February 21, 2007
  Loquendo and PIKA Technologies Inc., a designer and manufacturer of plug-in media processing hardware and software building blocks, announced the certified interoperability of Loquendo Speech Technologies and PIKA 's Digital T1/E1 PrimeNet MM boards, via Paraxip PSTN-to-SIP Gateway software.
  Through this partnership with Loquendo, PIKA enhances its position in the rapidly expanding market of automated voice applications for small and medium size businesses with one of the most advanced, state of the art speech recognition and text to speech technologies available today. Supporting 20 languages, Users of Loquendo technology will also be able to benefit from their VoiceXML browser capabilities.
  Loquendo's clients can now take advantage of the powerful PIKA Digital T1/E1 Boards, which provide superior voice processing, echo cancellation and speech detection technologies and are available in single, dual and quad T1/E1 port densities.
  >> More
  
  

Clients & Partners

• Loquendo And Magneti Marelli Announce Collaboration To Deliver Multimedia Navigators With Loquendo Text To Speech
  April 19, 2007
  Loquendo and Magneti Marelli, international leader in high-tech components and systems for the automotive industry, announced the integration of Loquendo's multi-lingual Automotive Text-to-Speech Solution into Magneti Marelli's latest car navigator, the MMSE 4G Navigation and Full Multimedia platform.
  The Magneti Marelli MMSE 4G Navigation and Full Multimedia platform has been installed by market leaders in Europe and Asia to offer their customers the latest in-car telematic technology, enabling Internet connection, rapid access to breakdown and emergency services and one of the most sophisticated satellite navigation systems available.
  >> More
  
  
• Loquendo Text to Speech Integrated into Activa Multimedia's interactive Virtual Characters
  April 12, 2007
  Loquendo, leading provider of speech technology, and Activa Multimedia, specialized in the creation of multimedia technologies and digital content, announce the integration of Loquendo TTS into Activa Multimedia's virtual character generator, known as iVAC (interactive Virtual Automatic Characters).
  Activa Multimedia's iVAC product line generates multimedia output based on virtual animated characters. The system enables the automatic production of video for TV, Internet and any mobile device. Activa Multimedia's characters are now able to speak using any of Loquendo's highly expressive, natural sounding voices, making them ideally suited for web, mobile or TV applications.
  >> More
  
  
• Loquendo and Artificial Solutions Announce Integration of their Technologies into IKEA Online Assistant
  March 27, 2007
  Loquendo and Artificial Solutions, specialist in optimizing and automating Customer Support through natural language solutions, announce the integration of Loquendo Text to Speech into Artificial Solution's Interactive Web Assistant for IKEA's UK website.
  The interactive customer care service on IKEA's website - Ask Anna - is a human-like online assistant that can move, smile, understand and now also speak using Loquendo's truly lifelike Text To Speech, answering customer queries and helping them to find their way around the site.
  Employing an Interactive Web Assistant for first line customer support, i.e. for responding to straightforward and repetitive questions, is highly cost efficient. Artificial Solutions' Interactive Web Assistant is particularly effective because it takes visitors straight to the appropriate webpage.
  >> More
  
  
• Loquendo TTS Installed in over 40 Major Taxi Companies in the UK Using the Autocab Taxi Booking System Powered by GPC Computer Software
  March 22, 2007
  Loquendo and GPC Computer Software, the largest manufacturer of taxi management and dispatch systems in the world, today announced the integration of Loquendo speech technology into GPC Computer Software's taxi operator solution, Autocab.
  Autocab (www.autocab.com) is the world's most popular booking and dispatch system, using proven and reliable computer and telecommunications technology to improve every aspect of the taxi operator's business, from booking and dispatch through to accounts and management. It has been installed in hundreds of locations throughout the world.
  >> More
  
  
• New Olidata Reglo Voice: the fully voice-controlled media center
  March 16, 2007
  Olidata, Italy's leading manufacturer of personal computers, and Loquendo announce the launch of a ground-breaking and easy to use voice-controlled media center: Olidata Reglo Voice. The solution has grown out of a partnership between Loquendo, providing the speech technology, and MOAI, providing the intelligent interface 'Speaky' - including remote-control unit with microphone and Push to Talk.
  Olidata Reglo Voice Media Center is able to voice control any menu: through a user friendly interface enriched with dynamic, TTS-based feedback users can change TV channel, make a content-based search of the TV or video guide or program the video recorder. Additionally, they can select a piece of music, view their photos, access any multimedia file they have saved in the media centre, or manage, by voice, VoIP telephony.
  >> More
  
  
• Silver Bullet Solution Brings Interactive Speech Applications Rapidly and Easily to Market like Never Before
  February 22, 2007
  Loquendo, SpeechVillage and Voxpilot launch the Silver Bullet solution, a milestone for the speech technology market. The Silver Bullet solution enables service developers and application designers to bring interactive voice services to market faster than ever before and at greatly reduced costs.
  Silver Bullet is the outcome of a close collaboration between three providers of state of the art technologies: the VoiceXML IVR Server from Voxpilot, the outstanding Speech Recognition and Text-To-Speech technology from Loquendo, and Speechdraw, the most advanced development tool on the market, from Speechvillage.
  Silver Bullet leverages proven open standards such as VoiceXML and MRCP to deliver a high quality, powerful and flexible solution. Silver Bullet is available in configurations of 12 and 24 ports in a single box and is expandable to thousands of ports. Designed with both the short and long-term in mind, Silver Bullet supports both ISDN and Voice Over IP (SIP) telecom interconnects. The integrated AudioCodes gateway provides E1/T1 connectivity to the PSTN.
  >> More
  
  

EVENTS

• Foro Comercial de Aplicaciones de Reconocimiento de Voz ,
  Centro Banamex, Mexico City, Mexico, April 24, 2007 (9:30 a.m. - 3:00 p.m.).
  
  
• EXPOCONTACT 2007,
  HOTEL MIRASIERRA SUITES, c/ Alfredo Marqueríe, 43 - Madrid, Spain, May 17, 2007.
  
  
• NAVIGATION EUROPE 2007,
  Marriott Hotel, Amsterdam, the Netherlands, 6-7 June 2007.
  
  
• The 7th ANNUAL LONDON CONNECTS CONFERENCE 2007,
  Conference Centre, QEII, London, UK, June 26, 2007.
  
  
• SpeechTEK 2007,
  Marriott Marquis, New York, BOOTH #509, 20-23 August 2007.
  
  
• CANALYS NAVIGATION FORUM 2007,
  The Grand Marina Hotel, Barcelona, Spain, September 10-12, 2007.