Hi,
I have successfully integrate the Pocket sphinx in my app . In Indian pronounce some words are not recognized by the app. Is there any option to training the Pocket Sphinx/ CMU Sphinx initially with user voice, then it works correctly?
" The adaptation process takes transcribed data and improves the model you already have. It's more robust than training and could lead to a good results even if your adaptation data is small. For example, it's enough to have 5 minutes of speech to significantly improve the dictation accuracy by adaptation to the particular speaker. "
Any example or method's available for user (speaker) to speech and add it into app ?
" The actual set of sentences you use is somewhat arbitrary, but ideally it should have good coverage of the most frequently used words or phonemes in the set of sentences or the type of text you want to recognize. For example, if you want to recognize isolated commands, you need tor record them. If you want to recognize dictation, you need to record full sentences. For simple voice adaptation we have had good results simply using sentences from the CMU ARCTIC text-to-speech databases. To that effect, here are the first 20 sentences from ARCTIC, a fileids file, and a transcription file "
Or else initially created the set of words or 20 sentences, and ask the user to speak first , all words are correctly recognized by app (80%) , then enter into main screen. In this process i acheive the Adaptive acoustic model ? .
Thanks ,
Saravanan.S
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
I have successfully integrate the Pocket sphinx in my app . In Indian pronounce some words are not recognized by the app. Is there any option to training the Pocket Sphinx/ CMU Sphinx initially with user voice, then it works correctly?
I was read the Adapting the default acoustic model , in http://cmusphinx.sourceforge.net/wiki/tutorialadapt this link ,
" The adaptation process takes transcribed data and improves the model you already have. It's more robust than training and could lead to a good results even if your adaptation data is small. For example, it's enough to have 5 minutes of speech to significantly improve the dictation accuracy by adaptation to the particular speaker. "
Any example or method's available for user (speaker) to speech and add it into app ?
" The actual set of sentences you use is somewhat arbitrary, but ideally it should have good coverage of the most frequently used words or phonemes in the set of sentences or the type of text you want to recognize. For example, if you want to recognize isolated commands, you need tor record them. If you want to recognize dictation, you need to record full sentences. For simple voice adaptation we have had good results simply using sentences from the CMU ARCTIC text-to-speech databases. To that effect, here are the first 20 sentences from ARCTIC, a fileids file, and a transcription file "
Or else initially created the set of words or 20 sentences, and ask the user to speak first , all words are correctly recognized by app (80%) , then enter into main screen. In this process i acheive the Adaptive acoustic model ? .
Thanks ,
Saravanan.S
Training is not supported on the phone. You have to send data to the server, update there and send the model back.
Thanks Nickolay .... :)
What data I have to send to the server ? Any server address is available .
I have to send my Dictionary and grammer file is enough ?
Please share the server link ..
You have to write the server code too.