IN-DEPTH

Loquendo would like to take this opportunity to applaud the efforts of the W3C, which recently published new standards that enhance the capabilities and interoperability of voice browsers and speech recognition systems. The W3C Voice Browser Working Group has completed work on both VoiceXML 2.1 and Semantic Interpretation for Speech Recognition (SISR) 1.0, two critical pieces of W3C's Speech Interface Framework.
This article describes the context in which these specifications were created and provides a description of the benefits they offer to the speech application market.

>> Go to the article.

Loquendo News

• Loquendo Expands. Company Grows its Sales and Marketing Structure
  July 30, 2007
  Loquendo announces the expansion of the sales force and the creation of new departments to cover the various geographical and business areas. The change is aimed at maximizing market impact through a more sophisticated product range and specialized sales force.
  >> More

STANDARDS CORNER

• VoiceXML 2.1 and SISR 1.0 are now W3C Recommendations
  June 19, 2007
  The World Wide Web Consortium (W3C) published two standards that enhance the capabilities and interoperability of voice browsers and speech recognition systems. The W3C Voice Browser Working Group has completed work on both: VoiceXML 2.1 and Semantic Interpretation for Speech Recognition (SISR) 1.0, two critical pieces of W3C's Speech Interface Framework. Read the W3C Press release and the Testimonials in support of the event from Aspect, Genesys, Loquendo, Nuance, VoiceXML Forum, Voxeo, and Voxpilot.
  
  
• VoiceXML Forum Launches VoiceXML 2.1 Platform Certification Program
  June 19, 2007
  The VoiceXML Forum, a global industry organization chartered to promote and to accelerate the adoption of VoiceXML-based applications, announced major enhancements to its Platform Certification Program. A pilot program, open to VoiceXML Forum members who have previously certified their platforms, began in June and will include certification testing for VoiceXML 2.1, SISR 1.0 and key features of other languages in the W3C Speech Interface Framework. The announcement was made in support of the announcement by the World Wide Web Consortium (W3C) that VoiceXML 2.1 and SISR 1.0 have been published as Web standards. To read more information about the VoiceXML Forum's Platform Certification Program, click here.
  
  
• Voice Browser and Multimodal Interaction meetings in Ca' Tron, Treviso
  June 18-22, 2007
  The W3C working groups of Voice Browser (VBWG) and Multimodal Interaction (MMIWG) were kindly hosted by H-care in Tenuta Ca' Tron at Roncade, Treviso, Italy. About 25 people were present from different countries to present the current status of the specification under development, and to discuss the most important issues. For instance in the MMIWG meeting the topics were MMI architecture, EMMA and InkML specifications and Multimodal Authoring, while in the VBWG the progress of SSML 1.1 and PLS 1.0 specifications and VoiceXML 3.0 were discussed. The central day of the week was a joint meeting where Umberto Basso, CEO of H-care, discussed their Human Digital Assistant, prof. Piero Cosi, CNR Padova, outlined his work on 3D Facial Animation and prof. Giuseppe Riccardi, DIT Univ. Trento, discussed his work on third generation conversational interfaces. All those present praised the beautiful location and the outstanding organisation and support of H-care's staff and the Fondazione Cassamarca. Public reports will be released shortly in the public archives of www-voice@w3.org and www-multimodal@w3.org.
  
  

R&D COLLABORATIONS

• Loquendo ASR R&D in DIVINES IST project - Diagnostic and Intrinsic Variabilities in Natural Speech
  In January 2007, the three year FP6 DIVINES project was successfully concluded, creating new acoustic-modelling and lexical modelling techniques as an alternative to current approaches. The project made significant contributions to the long-term goal of bridging the gap in speech recognition performance between human and machine.
  The techniques developed by the project achieved an improved capacity to handle the variability and non-stationarity that is intrinsic in speech: first, by using more appropriate adapted models (both acoustic and lexical) to improve voice recognition, and second, by accurate characterization of the phenomenon allowing the design of more robust or adaptable features/models and hence improvements in recognition accuracy.
  The consortium involved major European research centres, University departments and enterprises from Belgium, France, Germany and Italy, as well as leading voice technology R&D departments. It was composed of 8 partners: MULTITEL (Belgium), France Telecom (France), Carl von Ossietzky Universitaet Oldenburg (Germany), Babel Technologies S.A.(Belgium), Loquendo (Italy), Université d'Avignon et du Pays-vaucluse (France), Institut Eurecom - France, Politecnico di Torino (Italy).
  >> More
  
  
• Loquendo and the SHARE project at SpeechTEK 2007 conference in NY
  August 20-23, 2007
  At the SpeechTEK Conference, the Dortmund Fire Fighters and Loquendo will describe the expertise gained during the SHARE IST project (a joint European consortium funded by the European Community) using a Push-To-Share service. The project combines advanced, innovative technologies to allow emergency teams on-the-move, during large-scale rescue operations, to communicate naturally and bi-directionally, and to share structured multimodal information resources including audio, video, text, graphics and location information.
  The SHARE project will be present at SpeechTEK 2007 with a speech on: "Speech Technology in Advanced Multimodal Mobile Service for the Dortmund Fire Fighters' Emergency Forces" by Klaus Schäfer (Chief Fire Officer, Fire Department of Dortmund ), prof. Rainer Koch (University of Pederborn / Fire Department of Dortmund), and Davide Bonardo (Loquendo).
  
  To find out more about the SHARE project, click here. For general information about SpeechTEK, click here.
  
  
• LUNA project at Interspeech 2007
  August 27, 2007
  The LUNA project will be presented in a Tutorial given by prof. Giuseppe Riccardi and Sebastian Varges (University of Trento) at Interspeech 2007 on August 27, 2007.
  LUNA is a three-year project, supported by the European Commission, whose aim is to improve current automated telephone systems allowing easy human-machine interactions through spontaneous and unconstrained speech.
  
  For general information about the Tutorial at Interspeech 2007, click here. To find out more about the LUNA project, click here.
  
  
• Loquendo and the HIWIRE Project at Interspeech 2007
  August 30, 2007
  HIWIRE project will be presented at the special session "Novel techniques for the NATO non-native Air Traffic Control and HIWIRE cockpit databases" at Interspeech 2007 on August 30, 2007.
  Loquendo provided its ASR technology for the three year FP6 EU-funded project HIWIRE (Human Input that Works In Real Environments). The project has achieved significant improvements in ASR performance in noisy environments and for non-native speakers making considerable improvements to the robustness, naturalness and flexibility of vocal interaction between humans and machines.
  
  To read more about the special session at Interspeech 2007, click here. To learn about the HIWIRE project, click here.
  
  
• Loquendo and the SHARE project at the ISCA Interspeech 2007 conference in Antwerp - Belgium
  August 31, 2007
  The speech synthesis R&D group will present some results regarding the new features developed to enhance synthetic speech in noisy environments. This work was undertaken within the SHARE project whose goal is to develop a multimodal communication system supporting rescue operations and disaster management. Signal processing features like dynamic range control, gain and speech rate control were considered to enhance speech signals in outdoor and noisy environments. To this end, a particular configuration for controlling the signal dynamic has been investigated and assessed by means of subjective perceptual tests.
  
  The paper "Speech Synthesis enhancement in noisy environments" by Davide Bonardo(Loquendo) and Enrico Zovato (Loquendo) will be presented at Interspeech 2007 in the poster session "Speech Synthesis II" on August 31, 2007 (13:30-15:30).
  
  

LOQUENDO Products

• Loquendo TTS Powers Symbian OS™ V9 Operating System
  July 27, 2007
  Loquendo announced its latest development in high-quality, multilingual synthetic speech: Loquendo TTS has been ported to the Symbian OS™ v9 Operating System, which is the major new release from Symbian OS™ for smartphones, which boasts enhanced security, reliability and performance features.
  The new features of Loquendo embedded TTS for Symbian OS™ v9 include the same powerful text analysis capability as the server version alongside accurate semantic analysis and lexicon file customisation. Loquendo embedded TTS for Symbian OS™ provides full multilingual capability, supporting the TeleAtlas® and Navteq™ SAMPA phonetic alphabets, thus increasing the quality of speech rendering when reading out addresses, signage and points of interest - in any language - leveraging Loquendo TTS mixed language capability.
  With Loquendo embedded TTS for Symbian OS™ v9, users enjoy the benefits of Loquendo's emotional synthetic voices, which enable expressive content such as greetings, apologies, etc., to sound truly lifelike, for an even more natural effect.
  >> More
  
  
• The Voices of Charlotte and Olivier Consolidate Loquendo's Presence in the North-American Market
  June 14, 2007
  Loquendo announced the launch of Olivier and Charlotte, the brand new Canadian French voices for Loquendo TTS. These two new personas complement Loquendo's existing French-speaking voices - Juliette, Florence, Bernard and Sophie - enabling the company to support its partners ever more effectively in the North American markets, and in the larger part of the French speaking world.
  The new members of the Loquendo family bring the total number of languages in the Loquendo TTS portfolio to 20, with the total number of voices now at 50.
  To meet Olivier, click here and to meet Charlotte click here.
  >> More
  
  

Clients & Partners

• Ontario Systems Selects Loquendo TTS for its Voice Messaging Solution
  July 31, 2007
  Ontario Systems, LLC., a technology leader in debt collection software solutions, is pleased to announce an agreement with Loquendo to power Ontario Systems' voice messaging solution, Verified Contacts®, with Loquendo's innovative, multilingual speech technology.
  The Verified Contacts solution from Ontario Systems makes thousands of automated outbound calls, utilizing Loquendo speech technology to personalize calls, addressing each customer by name. Text to speech is key in voice messaging campaigns and it has proven to double right-party contacts. In addition, Ontario Systems' clients can use this technology to power staffed dialer campaigns with FOTI-compliant answering machine messages tailored with account specific data.
  >> More
  
  
• Travel Club's Online Booking Service Leverages Loquendo Speech Technology
  July 10, 2007
  Loquendo announced the selection of Loquendo speech technology for Travel Club's pioneering ticket-booking voice portal. More than 1,000 members currently fly each day with Travel Club, and the company has over 9 million members. An automated customer care service has now been developed by Telefónica to meet the needs of this expanding membership, incorporating Loquendo's award-winning text to speech technology.
  Travel Club's automated customer service is fully equipped to deal with all kinds of enquiries, including complex booking transactions, and provides a flexible service that callers feel comfortable interacting with, thanks to Loquendo's expressive synthetic speech and highly robust speech recognition, as well as to Telefónica's expertly designed automated dialogues.
  >> More
  
  
• Auronix Choose Loquendo's American Spanish TTS for Credito Real's Automated Call Services
  June 5, 2007
  Loquendo and Auronix, a leading provider of computer telephony applications, announced an innovative telephony application developed by Auronix in Mexico. The solution, which incorporates the Loquendo TTS speech engine, is successfully used by Credito Real throughout Mexico.
  With hundreds of thousands of customers, Credito Real is the premier provider of customer finance for Mexico's major furniture stores, and has been using Auronix technology for more than 8 years. Credito Real recently upgraded their Aurodialer call-collection system to include Loquendo's American Spanish TTS voice, the expressive and natural-sounding Carlos. This upgrade yielded such a significant increase in collection rates for Credito Real, that a second system was immediately deployed.
  >> More
  
  
• Cystelcom Adds Loquendo TTS Technology to Veedia Multimedia Platform
  May 17, 2007
  Loquendo and Cystelcom, provider of solutions for contact centres and 3G telephony, announced the integration of Loquendo text to speech into Cystelcom's Veedia™ platform for mobile video solutions.
  The Cystelcom Systems Veedia™ platform, distributed in Spain by Multichannel Technologies, provides a complete range of applications for service providers, mobile operators and companies offering interactive video telephony services for 3G users and broadband networks supporting IP.
  With the Veedia™ platform, operators can launch a wide range of 3G services with minimal time to market. 3G phone users can then access the latest video applications, both live and pre-recorded: customers can connect with other users, send and receive video messages, access video newspapers, monitor closed circuit camera systems, etc., all leveraging on Loquendo's highly intelligible and human-like text to speech.
  >> More
  
  
• Loquendo And BySide.Com Announce A Partnership To Deliver Innovative Contact Center And CRM Applications
  May 8, 2007
  Loquendo and BySide.Com, the Web Contact Centre and Web Customer Relationship Management (CRM) company, announce the integration of Loquendo speech technology into BySide WebCare. This partnership enables BySide.Com to offer to its clients new and innovative technology, such as virtual call centres and self service applications.
  BySide WebCare is an automated, online, telephony solution which, thanks to Loquendo speech technology and the Loquendo VoiceXML/CCXML platform, boasts greatly enhanced performance and efficiency. The Loquendo platform enables telcos, businesses, contact centres and emerging operators (such as VoIP) to develop voice-enabled applications and solutions for any Web architecture. The solution provides integrators with the latest features of Loquendo's acclaimed speech recognition and text-to-speech technology, with complete support for VoiceXML and CCXML standards, which allow users to make the most of speech applications. CCXML makes call control more flexible, while VoiceXML focuses on the interactive aspects of speech applications.
  >> More
  
  

EVENTS

• SpeechTEK 2007,
  Marriott Marquis, New York, 20-23 August 2007
  Visit Loquendo at BOOTH #509.
  
  
  • Monday, August 20, 2007
    TRACK C - ADVANCED SPEECH TECHNOLOGIES SYMPOSIUM
    C102 - Applying the Pronunciation Lexicon Specification to ASR & TTS
    11:30 am - 12:30 pm (Patrizio Bergallo, Loquendo)
    
    
  • Tuesday, August 21, 2007
    TRACK D: INNOVATIVE SPEECH APPLICATIONS
    D201 - Speech Technology in Advanced Multimodal Mobile Service for the Dortmund Fire Fighters' Emergency Forces
    10:45 am - 12:00 pm (Klaus Schäfer, Fire Department of Dortmund, Germany, with Davide Bonardo, Loquendo)
    
    
  • Tuesday, August 21, 2007
    TRACK E - TIPS & TECHNIQUES FOR IMPLEMENTING YOUR SPEECH APPLICATIONS
    E201 - New Applications & Contexts for Speech Technology
    1:30 pm - 2:30 pm (Davide Bonardo, Loquendo)
    
    
  
  
• Interspeech 2007,
  Flanders Congress and Concert Centre, Antwerp, Belgium, August 27-31, 2007.
  
  
  • Wednesday, August 29, 2007
    WeD.O1-4: Speaker verification & identification III
    Loquendo - Politecnico di Torino's 2006 NIST Speaker Recognition Evaluation System
    17:00 pm (Claudio Vair, Loquendo)
    
    
  • Thursday, August 30, 2007
    ThD.SS: Novel techniques for the NATO non-native air-traffic control and HIWIRE cockpit databases
    Experiments on Hiwire database using Denoising and Adaptation with an hybrid HMM-ANN Model
    17:00 pm - 18:00 pm (Franco Mana, Loquendo)
    
    
  • Friday, August 31, 2007
    FrC.P2-2: Speech synthesis II (poster)
    Speech Synthesis enhancement in noisy environments
    13:30 pm - 15:30 pm (Enrico Zovato, Loquendo)
    
    
  
  
• CANALYS NAVIGATION FORUM 2007,
  The Grand Marina Hotel, Barcelona, Spain, September 10-12, 2007.
  
  
• Call Centre Expo 2007,
  Birmingham NEC, UK, September 26-27, 2007.
  
  
• VoiceDays 2007,
  Bonn, Alter Bundestag, Germany, October 17-18, 2007.
  
  
• Navigation USA 2007,
  San Francisco Bay, USA, November 29-30, 2007.