Speech Recognition

We currently use CMU Sphinx for all of our voice recognition tasks.

Sphinx is open source voice recognition software available from the CMU speech group. The software is designed to translate voice input into text. The system first uses Sphinx to recognize the words of the speaker, which are in turn passed on to the discussion engine to be processed for a response. Sphinx operates through the use of language models to understand how textual words are pronounced and which words are commonly joined.

Examples

QuickTime (1.3M)
Media Player (8.2M)