Edit Message

Re: Create script to train up acoustic models on speech audio from Librivox project.

Well - who am I to disagree?

One thought I've just had - I've heard of speech recognition companies who have trained up models using HTK and run them under SPHINX (before  HDecode was available).   Surely it can't be that hard, well certainly the means/variances/transition probs should be trivial.  Voxforge would seem to be an excellent place to:

1) develop some GPL code to do the acuostic model conversion (lets assume ARPA format LMs)

2) benchmark the ISIP/HTK/Julius/Sphinx recognisers using exactly the same acoustic models, pronunciations and language models.

Anyone got any results to share?

 Tony