Menu

Sphinx Post processing tool language model and accuracy

Help
2017-12-06
2017-12-07
  • Maathangi Sankar

    Hi!
    I'm interested in adding appropriate punctuations to the transcribed ASR output for English. I used the post-processing framework as part of Sphinx using the Gutenberg lm model.
    https://cmusphinx.github.io/2012/08/postprocessing-framework/
    Wondering if there's an update to this language model or this branch that I can use for better results?
    When I tried this for a passage from the Gutenberg text corpus, it appears that after some initial phrases, commas are getting added ib between every word. Any idea why this might be happening or pointers to what I can do to improve the accuracy here?

    Any help regarding this would be super awesome!
    Thank you very much!

     
    • Nickolay V. Shmyrev

       
  • Maathangi Sankar

    Thank you very much!!

     

Log in to post a comment.

Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.