This is Sphinx 3.7 (s3.7), one of Carnegie Mellon University's open
source large vocabulary, speaker-independent continuous speech
recognition engine. Please see the LICENSE file for terms of use.
There has always been confusion of the versioning of Sphinx-3. Here
is a brief description.
Over the course of development, Sphinx-3 had branched into two major
versions. They are the flat lexicon decoder, (or aka "s3-slow") and
the tree lexicon decoder (or aka "s3-fast"). Performance wise,
s3-slow is ~10% relative better than s3-fast empirically. However,
s3-fast could be around 10 times faster than s3-slow. From sphinx 3.0
to Sphinx 3.4. The two code-bases were separated.
During the development of Sphinx 3.5, conscious decision were made to
merge the two branches together. This merging is stabilized in Sphinx
3.6. Internally, both the fast decoder and slow decoder share over
70% of the codebase. From the user perspective though, the old
sphinx3_decode_anytopo (s3-slow) and sphinx3_decode (s3-fast) exist as
two separate interface of the decoding engine. This will allow
backward compatibility to the past.
In 3.7, we did a site-wide change that allow sphinx3 and pocketsphinx
(and perhaps later SphinxTrain) to share the same codebase.
THIS IS A RESEARCH SYSTEM. After several extensions and upgrade, we
still find that Sphinx is better regarded as an early release of a
research system. We know the APIs and function names are likely to
change, and that several tools need to be made available to make this
all complete. With your help and contributions, this can progress in
response to the needs and patches provided.
If you install the distribution in the default location, you can
access the documentation here:
file://localhost/usr/local/share/sphinx3/doc/
There are couple of useful documents you could find in the ./doc
directory. These include:
s3_codework.html : Code review document prepared by Ravi Mosur.
s3_description.html : Code review document prepared by Ravi Mosur.
s3_overview.html : Code review document prepared by Ravi Mosur.
s3.2.ppt : The powerpoint presentation prepared by Ravi Mosur.
FAQ.html : Frequently asked question page prepared by Rita Singh
sphinxman_manual.html : Manual for training prepared by Rita Singh.
sphinxman_FAQ.html : FAQ for training prepared by Rita Singh.
sphinxman_misc.html : Miscellaneous information for training prepared by Rita Singh.
models.html : Description of the default broad cast news model
(Obsolete) cmdhelp.txt : Sphinx 3.3 command line arguments.
For up-to-date information, please see the web site at
http://cmusphinx.org
A brief note of the API: From s3.5, the live mode API of the decoder
is stabilized. The decoder routine is stored in a library called
libs3decoder.a. The decoder function takes blocks of speech samples
and returns partial hypotheses for the blocks decoded during
runtime. Examples of its use are given in main_live_example.c and
main_live_pretend.c
This directory contains the scripts and instructions necessary for
decoding speech input using the CMU Sphinx Recognizer.
This distribution is free software, see LICENSE for license.
Installation Guide:
^^^^^^^^^^^^^^^^^^^
This sections contain installation guide for various platforms.
Linux/Unix Installation:
^^^^^^^^^^^^^^^^^^^^^^^^
Starting from Sphinx 3.7, you need to compile sphinxbase before
compilation of sphinx3. The layout of the directory should look
like this:
sphinx3/
sphinxbase/
By default, sphinx3 will search for ../sphinxbase/ as the path of the
include files and libraries. You need to compile sphinxbase FIRST
before compiling sphinx3. Please read the README in the sphinxbase
package for details.
This distribution now uses GNU autoconf to find out basic information
about your system, and should compile on most Unix and Unix-like
systems, and certainly on Linux.
If you downloaded directly from the Subversion repository, you need to
create the "configure" file
>./autogen.sh
>./autogen.sh
(Make sure you do this twice.)
If you downloaded a release version, or if you have already run
"autogen.sh", you can build simply by running
>./configure
>make
(Or whatever you call GNU Make). This should configure everything
automatically. After a successful compilation, you may test the system
by running
>make check
and then install it with
>make install
This defaults to installing Sphinx-3 under /usr/local. You may
customize it by running ./configure with an argument, as in
>./configure --prefix=/my/own/installation/directory
Windows Installation:
^^^^^^^^^^^^^^^^^^^^^
Starting from sphinx 3.7, you need to install sphinxbase before
installation of sphinx. The general layout looks like this
sphinx3/
sphinxbase/
To compile the sphinx 3.7 in Visual C++ 6.0,
1, unzip the file.
2, click sphinx3.dsw
3, build the application projects by clicking "Build" at the tool bar.
Step 3 can be done easily in VC6.0 by clicking the "Build" menu at
the top bar, and selecting "Batch Build...". Make sure all projects
are selected, preferably the "Release" version of each. You can also
run with the "Debug" version, but this version is usually slower. To
run the executable, please remember to set the corresponding active
project.
If you are using cygwin, the installation procedure is very similar to
the Unix installation. However, there is no audio driver support in
cygwin currently so one can only use the batch mode recognzier.
Acknowldegments
^^^^^^^^^^^^^^^
This work was built over a long years at CMU by most of the people in
the Sphinx Group. Some code goes back to 1986. The most recent wor for
release after sphinx 3.0 includes the following, listed
alphabetically.
Alan W Black (awb@cs.cmu.edu)
Arthur Chan (archan@cs.cmu.edu)
Evandro Gouvea (egouvea+@cs.cmu.edu)
Ricky Houghton (ricky.houghton@cs.cmu.edu)
David Huggins-Daines (dhuggins@cs.cmu.edu)
Kevin Lenzo (lenzo@cs.cmu.edu)
Ravi Mosur
Rita Singh (rsingh+@cs.cmu.edu)
Yitao Sun (yitao@cs.cmu.edu)
Eric Thayer
