Language Model Creation

Streamline what exactly ?

similar to the change we did in taskcluster.py, ideally, one should not juggle with handling different versions of files manually, ideally.

I don’t see the link with the current thread, can you elaborate ?

I see, I’ll try to come back with a more concrete example. Thanks for the great work. We have about 80% accuracy thanks to you. :slight_smile:


it is fixed. Google drive import had some problems… (bow)
words.arpa
lm.binary and trie is generated successfully. hooray

1 Like