Hey there,
I'm completely new to Speech Recognition and just tried out the PocketSphinx API for Android.
I'm trying to recognize some (~20) keywords in a text that comes from a bluetooth headset. The English 8khz acoustic model works fine, but the app needs to support German.
So what is the best way to generate a German 8khz acoustic model? Can I adapt the English 8khz model or the German 16khz model or do I have to train a completly new model? Or do you maybe have an acoustic model that I could use?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hey there,
I'm completely new to Speech Recognition and just tried out the PocketSphinx API for Android.
I'm trying to recognize some (~20) keywords in a text that comes from a bluetooth headset. The English 8khz acoustic model works fine, but the app needs to support German.
So what is the best way to generate a German 8khz acoustic model? Can I adapt the English 8khz model or the German 16khz model or do I have to train a completly new model? Or do you maybe have an acoustic model that I could use?
You can contact Guenter Bartsch guenter.bartsch@gmail.com, they guy who creates German models and ask him to create a model for you.
Oh thank you for your answer! I'm going to mail Guenter Bartsch then.