Loquendo, global speech technology provider, has released Loquendo ASR 7.10, now supporting speech recognition based on more complex statistical language models (built on N-grams where "N" can be greater than 3), enabling the creation of advanced speech applications (i.e. for call routing and transcription).

As a result, Loquendo ASR now recognizes more linguistically complex utterances. The new version, licensed under Tier 6, also increases efficiency when applied to language models using bigrams and trigrams.

For grammar-based speech recognition applications, ASR 7.10 now introduces a new capability: grammars can be modularized into a fixed part (e.g. hundreds of thousands of words that are infrequently modified) and a dynamic part containing new entries. When a new grammar is created, it can reference a secondary and pre-compiled grammar that the system then loads dynamically. In this way, a large grammar containing many thousands of names can be loaded rapidly at runtime; offline, the grammar can then be modified or completely recompiled as needed without having to modify the primary grammar.

This feature is particularly useful, for example, for grammars containing lists of company personnel or stocks and shares. The new feature can also be adopted in command-and-control scenarios, enabling, for instance, voice-controlled media players to handle recently uploaded tracks (which could be loaded together with pre-compiled grammars).

Loquendo's speaker-independent ASR comes with various acoustic models specialized for different application fields, such as telephony (fixed, mobile, VoIP) and noisy environments (in-car). It is also available in compact versions for embedded devices.

Like all Loquendo technologies, Loquendo ASR is available for both Windows and Linux, and for automotive and smartphone platforms. Follow these links for all supported Operating Systems, Standards and available Languages.

About Loquendo

Named Speech Technology's Market Leader for the past five consecutive years, Loquendo provides a complete range of speech technologies for telephony, mobile, automotive, embeddedm and desktop solutions in 32 languages with 76 voices, and constantly growing.

Loquendo TTS and Loquendo ASR empower people to interact with technology in the most natural way possible – using their voice – creating a next-generation client experience while saving businesses millions each year. Also integrable via the Loquendo MRCP Server and VoxNauta VoiceXML, CCXML & SCXML platform, Loquendo speech technologies power millions of calls every day in the telecommunications and enterprise markets across the globe. Loquendo TTS and ASR are also deployed in more than 20 million devices worldwide in embedded, automotive and mobile environments. 

Loquendo is a Telecom Italia company headquartered in Turin, Italy, with offices in the US, UK, Spain, Germany and France, and a global network of partners. For more info, and to hear Loquendo TTS for yourself, go to www.loquendo.com