I am looking into speech recognition software and I would really like to use sphinx, it seems very powerful.
I really don't know too much about ASR in general. Is there a good tutorial on it?
Also I have tried to follow the Sphinx tutorial online and it works until I have to run the decode/slave.pl, there is no decode directory. I have also tried to run sphinx3-simple and it fails because it is looking for a livedecode application which doesn't exist, do I use the decode app here?
I am using the nightly tarball from 12/18/2005
Thanks!
Glenn
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've read the tutorial your talking about and my guess that the decoder has been removed, but you can still test your model using sphinx4. here is a tutrial on how to convert your model to a jar file (so sphinx4 can use it) http://cmusphinx.sourceforge.net/sphinx4/doc/UsingSphinxTrainModels.html once you have done that, assuming you have trained an4 audio set, you can run the an4 batch test included in sphinx4.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi All,
I am looking into speech recognition software and I would really like to use sphinx, it seems very powerful.
I really don't know too much about ASR in general. Is there a good tutorial on it?
Also I have tried to follow the Sphinx tutorial online and it works until I have to run the decode/slave.pl, there is no decode directory. I have also tried to run sphinx3-simple and it fails because it is looking for a livedecode application which doesn't exist, do I use the decode app here?
I am using the nightly tarball from 12/18/2005
Thanks!
Glenn
I've read the tutorial your talking about and my guess that the decoder has been removed, but you can still test your model using sphinx4. here is a tutrial on how to convert your model to a jar file (so sphinx4 can use it) http://cmusphinx.sourceforge.net/sphinx4/doc/UsingSphinxTrainModels.html once you have done that, assuming you have trained an4 audio set, you can run the an4 batch test included in sphinx4.