Sensory Releases Speech Recognition SDK for Linux
SANTA CLARA, CA - Sensory Inc., a supplier of embedded speech technologies, announced that it has ported its Fluent Speech Software to the Linux platform. Linux programmers can now add this to their PC based applications. Using a proprietary text based phonetic engine, the Fluent Speech SDK allows custom tuning of speech recognition sets containing thousands of words or phrases without the need to verbally train the computer. "The Fluent Speech SDK is an ideal platform for programming speech recognition into dynamic database applications, as it allows recognition sets to be generated on the fly from incoming text based data," said Bill Teasley, Sensory's vice president of engineering. "The flexibility and convenience of this approach is very powerful and a significant addition to the Linux community." Features like word spotting, phrase grammars and continuous digit recognition create a more natural user interface. The Fluent-Speech SDK is written in standard ANSI C language, and includes dynamic and static software libraries, modifiable example code, APIs for building and tuning the recognition engine, and also a demo program to help conceptualize the final product. The Fluent Speech SDK is available now, and costs $1,995.