systems, and data loggers.
Fonix says its proprietary neural network-based speech technology provides accurate speaker-independent recognition, even in noisy environments, and does not require the user to train the system to recognize his or her speech, or to a specific acoustic environment. The technology accomplishes high recognition rates without the need for intensive CPU or memory resources, and provides control functions that respond to continuous speech, enabling users to speak naturally, without pausing. It also supports nine languages -- US and UK English, Canadian and European French, Castilian and Latin American Spanish, German, Swedish, and Japanese.
According to Fonix, VoiceIn 4.0 SE comes with the following tools:
- Phonetic voice tags
- Wide characters for improved Asian language support
- Optimized alpha-digit capability
- Multi-pass recognition
- The ability to create recognition grammars dynamically during run-time
- Improved memory usage
- Faster processing speed
- Better out-of-vocabulary recognition
- A simplified API to make development faster and easier and to reduce time-to-market
VoiceIn 4.0 SE runs on ARM, Xscale, Renesas (SH3 and SH4), and MIPS processors.
Related stories: 
