Results from GSoC?

Speech Recognition Toolkit

Brought to you by: air, arthchan2003, awb, bhiksha, and 5 others

This project can now be found here.

Results from GSoC?

Forum: Help

Creator: Daniel Wolf

Created: 2017-10-11

Updated: 2017-10-11

Daniel Wolf - 2017-10-11

For the Google Summer of Code 2017, several related projects were started to add deep neural networks to PocketSphinx. I was thrilled when I read the news back in May, because improved recognition accuracy would help me a lot.

So I'd love to learn about the outcome! Will some of the results be integrated into PocketSphinx in the near future?

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
- Nickolay V. Shmyrev - 2017-10-13
  
  Created code is here:
  
  https://github.com/cmusphinx/sphinxtrain/pull/14/files
  https://github.com/shinigami1494/DNN-For-PocketSphinx
  
  It is semi-finished idea though, needs much more work.
  
  If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
  - Daniel Wolf - 2017-10-14
    
    Thanks for the links! I skimmed the repos but I'm afraid that my knowledge in machine learning is much too limited.
    
    Can you give any estimate if and when this will be integrated into PocketSphinx? (Are we talking six months, a year, ...?)
    
    And once it is integrated, will it add complexity to the build process (OpenCL etc), or will the code still be portable C?
    
    If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Log in to post a comment.