I'm planning on writing a program for Linux that uses text to speech and speech recognition. What are the best tools/libraries for this? Should I use Windows instead to be able to use better tools? The tools need to be easily callable from a console or C program.
A:
I know espeak is a very good text-to-speech program for linux (it can even do different accents!), but I don't know of any speech recognition systems designed for UNIX.
Rob Golding
2009-05-18 12:35:25
+2
A:
I have used both Loquendo and Festival under linux. I would consider the festival voices I have used pretty poor, with very robotic synthesis. The Loquendo voices, on the other hand, are excellent - very high quality.
mysomic
2009-05-18 13:15:17
If you are going to use Festival, you should install the alternate voices. Instructions (for debian/ubuntu) are here: http://ubuntuforums.org/showthread.php?t=677277
Matt G
2009-05-18 13:57:34
You just like leaving us hanging don't you? Of course there are what?
Chas. Owens
2009-05-18 14:04:51
How was your experience with Loquendo? If you're up for it, I'd love to ask you a couple questions about it by email?
philfreo
2010-07-17 07:54:21
+3
A:
For speech recognition there exists very little for linux. I were only aware of one apparently decent option, something IBM released some years ago but later was no longer made available (anyone knows if this ViaVoice SDK is still possible to get hold of from anywhere?). There are some more information about possible options at wikipedia.
hlovdal
2009-05-18 13:46:57
ViaVoice SDK. It was never in full release, and the docs demanded a fairly narrow range on 2.4 kernel release numbers. I toyed around with it for taking some typing load off when I had intermittent tedonitis in my wrists, but no luck...
dmckee
2009-05-18 14:53:10
+1
A:
For speech recognition there are the various Sphinxes. The different variants have different pros and cons, there is a comparison here Comparison of Sphinx versions. Sphinx 4 is Java, but the others are C, I believe.
Matt G
2009-05-18 13:53:57
+4
A:
It depends quite a bit on what speech you are trying to recognize.
This is an article from 2005 that explains some of the difficulties in creating a dictation program: http://www.cs.cmu.edu/~archan/personal/whyNoOpenSourceDictationDraft4.html . If you want that, the Julius speech recognition engine seems promising, but you will need to add your own acoustic and language models. You might be able to use the voxforge acoustic model.
If you are not trying to write a dictation program then you have a much easier task. Command programs have limited vocabularies, for example 'If you would like to continue in English, say "English"'.
I was able to get pretty good results using pocketsphinx and gstreamer to make a program that automatically edits most occurrences of the word "twitter" out of the TWiT podcast. It didn't work at all until I used my own language model based on transcripts of the podcast; the machine transcriptions from the speech recognizer are useless/hilarious but they do an okay job of finding the keyword.
joeforker
2009-05-18 14:01:00
do you have any experience with using pocketsphinx and gstreamer with the tcpserversrc/client?
blackkettle
2009-08-26 14:53:40
no, but gstpocketsphinx + tcpserversrc/sink shouldn't be any different than any other gstreamer element + tcpserversrc/sink.
joeforker
2009-08-26 15:06:51
@joeforker the link to your "twitterkiller" program appears to be broken.
Steven Oxley
2010-06-08 22:45:15
A:
at&t fsm toolkit is also pretty awesome - no commercial use allowed though,
blackkettle
2009-08-26 14:55:38
A:
This is a bit old but I saw that a fairly comprehensive guide on speech recognition on Hackaday a few days ago: http://hackaday.com/2010/07/09/get-started-with-speech-recognition/
Cory Walker
2010-07-11 22:24:35