CMU Open Source Sphinx-3 Decoder
Sphinx Speech Group
School of Computer Science
Carnegie Mellon University
Pittsburgh, PA 15213
This is the current working Sphinx-3, s3.3 code base. It contains
a front end that allows for stream processing. However s3.3 does not
split nor parse the incoming audio data. It assumes it has been
segmented already. The provided set of models requires input signal to
be sampled at 16kHz, and requires audio in raw PCM format.
For more information regarding Sphinx-3, please check the following
links
- An overview of the Sphinx-3
decoders (notice the plural!!!)
- Information about the models. You
must read this if you intend to successfully use the models
provided.
- Instructions if you are compiling
under MS Development Studio.
- Sphinx-3 decoder manual
- Sphinx-3 front-end module
- Sphinx FAQ in local HTML or remote HTML
(Remote promises to be more up to date.)
- Talk on the Sphinx-3 transition to Sphinx-3.2 in PDF or PPT format.
- CMU Open Source Sphinx web page: cmusphinx.org
- CMU Speech web page: www.speech.cs.cmu.edu
Maintained by Evandro B. Gouvêa
Last modified: Sun Nov 17 23:12:42 EST 2002