Hi,
The Sphinx4 works well rather to say perfectly for the US English. But Does not recognize the Indian English Accent. The dictionary which HUB4 uses does not include some of the phones like AX,EA,IA,OH.
So I have to change the whole phoneme list. Any idea how to proceed? Or is there any other way?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear prar, posting into multiple forums will not help you, please avoid it in the future. About your problem - you have to train your own model for your domain. If your task is not continuous speech recognition you can probably do it other way. Refer to Sphinxtrain docs for details:
Thanks Shmyrev for the guidance. It is continuous speech recognition, so I dont think I have any choice other than training.
And Sorry for posting in two forums!! Do not know how to delete a post! Will not repeat it again!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
The Sphinx4 works well rather to say perfectly for the US English. But Does not recognize the Indian English Accent. The dictionary which HUB4 uses does not include some of the phones like AX,EA,IA,OH.
So I have to change the whole phoneme list. Any idea how to proceed? Or is there any other way?
Dear prar, posting into multiple forums will not help you, please avoid it in the future. About your problem - you have to train your own model for your domain. If your task is not continuous speech recognition you can probably do it other way. Refer to Sphinxtrain docs for details:
http://www.speech.cs.cmu.edu/sphinxman/scriptman1.html
Thanks Shmyrev for the guidance. It is continuous speech recognition, so I dont think I have any choice other than training.
And Sorry for posting in two forums!! Do not know how to delete a post! Will not repeat it again!