Just playing with the toolkit and learning about speech recognition, so this is likely a truly newbie question, but...
What do folks use to generate the audio data files to feed into Sphinx? Like most folks I have a Windows machine handy. If I use the Sound Recorder that comes with Windows, what format do I save the sound file in? What formats does Sphinx take as input?
Thanks,
Jerry
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2001-10-21
Hmm. I see that answer a couple of posts down. :-)
So the next question is does anyone know if Windows' Sound Recorder prepends the usual WAV headers to the file it saves and if so, how do I strip it off to end up with a raw file?
Thanks,
Jerry
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Just playing with the toolkit and learning about speech recognition, so this is likely a truly newbie question, but...
What do folks use to generate the audio data files to feed into Sphinx? Like most folks I have a Windows machine handy. If I use the Sound Recorder that comes with Windows, what format do I save the sound file in? What formats does Sphinx take as input?
Thanks,
Jerry
Hmm. I see that answer a couple of posts down. :-)
So the next question is does anyone know if Windows' Sound Recorder prepends the usual WAV headers to the file it saves and if so, how do I strip it off to end up with a raw file?
Thanks,
Jerry
Usually wav header are 44 bytes long. Just
skip these bytes to find samples.
You can also convert from wav to raw pcm format
with cool edit 96. Search the web for cool96.exe
It looks like we should get WAV/RIFF support into the front end.