Turin, July 06th, 2005

Loquendo, the global speech technology company, announces the release of Loquendo ASR 6.7. The new version of Loquendo’s Automatic Speech Recognition gives customers the means of greatly improving engine performance by giving them the necessary tools to efficiently adapt their own Acoustic Models, effortlessly and without the need for specific technical know-how.

Systems integrators can now easily make the recognition engine adapt to difficult conditions such as way of speaking (i.e. regional accents, fast speaking), different audio channels (i.e. wireless, multimedia microphone, PDA, etc), different environments (i.e. in-car, outdoors, public areas, factory), application-dependent vocabulary (e.g. some specific jargon, like aeronautic terms). Acoustic model adaptation dramatically improves recognition performance, using audio material recorded online by Loquendo ASR while the application runs.

“Loquendo is excited to be able to provide its partners with all the necessary tools to create efficient applications by themselves, without the support of external speech experts.” - said Rosanna Duce, Vice President of Loquendo - “By exploiting Acoustic Model Adaptation, there is a significant increase in recognition performance in all those cases where the speech service has specific characteristics. An application that is repeatedly used by the same speaker can be improved if the acoustic models are adapted to him. In the same way, an application that is used in a particular environment (e.g. car) or using a certain kind of input device (e.g. a particular PDA or smart-phone) can be improved if the general models are adapted with vocal material recorded from the user interactions.”

Furthermore, Loquendo ASR 6.7 also provides a powerful tool to achieve a more effective and user-friendly use of speech recognition in voice application development. With Voice Enrolled Grammars, end-users can vocally customize personal services, including personal agenda and contact management in Voice Activated Dialling services.

Loquendo ASR is the next-generation speech recognition technology that makes it possible to create automated voice solutions that are accurate, flexible and robust - even in the noisiest of environments. The wide range of Loquendo ASR languages is available in Argentinean, Brazilian, Castilian Spanish, Catalan, Chilean, Dutch, French, German, Greek, Italian, Mexican, Portuguese, Swedish, U.K. and U.S. English.
Upon request, certain languages can be provided as a single module, e.g. Castilian and Catalan.

About Loquendo - Vocal Technology and Services
With over 30 years of R&D expertise, Loquendo is at the forefront of the global speech technologies market. Its high-quality, high-performance technologies (Loquendo TTS - text to speech, Loquendo ASR - automatic speech recognition, Loquendo Speech Server, Loquendo Speaker Verification) and platforms guarantee systems integrators the best solutions in 16 languages and 36 voices - with more coming soon.
From its headquarters in Turin, Italy, Loquendo deploys technologies that power millions of calls every day in the telecommunications and enterprise markets throughout the world.

For more information, visit www.loquendo.com.