Turin, Italy- based speech technologies provider Loquendo (News - Alert) has announced enhancements and a new release of its speech recognition engine.

 

According to officials with Loquendo version 7.8 of Loquendo ASR now includes the ability to export word lattices, improved recognition accuracy for British and English, and reduced time for building an Adapted Acoustic Model.

Through these new feature enhancements, it will be possible for users to gain more accurate recognition of natural, spontaneous speech and it’s now much faster to allow acoustic models to be adapted to a particular speaker or to a specific environment like an in-car or other noisy environment.

Loquendo ASR, which is currently available in 25 languages, can be used for both Windows and Linux. Also, Loquendo Embedded ASR supports all major embedded operating systems. The offering is used to power services like automated directory assistance, automotive applications and many others across the globe.

 

Recently, the company announced it was recognized as a “Market Leader” for their advanced speech technologies “accuracy, cost, customization, integration and innovation”

 

At an industry event, the company was presented with a ‘Speech Engine Market Leader’ award and Liquid’s director of International Standards, Paolo Baggia, was presented at with a ‘Speech Luminary 2009’ award.