Combine acoustic model with language model and detect silence between words

Acoustic Model Discussions

Flat

User: Head7
Date: 4/4/2017 5:46 am

Views: 6314
Rating: 0

Hello everybody , trying to build a SR system. How can i combine an acoustic model with a language model. i mean what are the inputs and outputs of each one?

And how to detect the beginning and end of words in continous speech?

--- (Edited on 4/4/2017 5:46 am [GMT-0500] by ) ---

Re: Combine acoustic model with language model and detect silence between words

User: Praiseworthy
Date: 4/10/2017 10:32 am

Views: 77
Rating: 1

Hi, which tool are you using for building you Speech Recognition System?

--- (Edited on 4/10/2017 4:32 pm [GMT+0100] by Praiseworthy) ---

Re: Combine acoustic model with language model and detect silence between words

User: Head7
Date: 4/11/2017 3:30 am

Views: 66
Rating: 0

I'm using python/tensorflow to implement a recurrent neural network for acoustic modeling

--- (Edited on 4/11/2017 3:30 am [GMT-0500] by Head7) ---

Re: Combine acoustic model with language model and detect silence between words

User: Praiseworthy
Date: 4/11/2017 8:47 am

Views: 2727
Rating: 1

Hi,

I think you need to read up on speech recognition, in order to know more about the general structure of it. Here are some links that could provide that. I Hope you find them useful.

http://www.cs.columbia.edu/~mcollins/6864/slides/asr.pdf

https://www.inf.ed.ac.uk/teaching/courses/asr/2012-13/asr01-intro-4up.pdf

https://www.youtube.com/watch?v=i9Gn2QYrYpo

https://www.youtube.com/watch?v=Wb3YXPlo0GA

http://stackoverflow.com/questions/12239080/getting-started-with-speech-recognition-and-python

--- (Edited on 4/11/2017 2:47 pm [GMT+0100] by Praiseworthy) ---

Previous • Next •


Username	Password