Acoustic Models

The Vestec Speech Engine has been designed to support speech recognition in all major spoken languages. In order to conduct speech recognition in a particular language, the developer needs to load the acoustic model for that language. The engine comes with American English acoustic model which is well suited for English-based applications in both United States and Canada.

Work is underway on acoustic models for other major European as well as Asian languages, including German, French, Spanish, Italian, Portuguese, Arabic, Chinese, and Hindi. In addition, we are developing acoustic models for different English accents, such as British English, Australian English, Indian English, and Singaporean English.

Unless stated otherwise, all acoustic models available at the time of engine purchase are included in the licensing fee. Acoustic models released after engine purchase are considered a software “upgrade” and can be generally downloaded upon payment of annual software maintenance fee. Some restrictions apply.