Trying to have the inference demo work - but no success
Closed this issue · 3 comments
olup commented
I am trying to have the inference demo output something intelligible in this collab.
But for now all my efforts are vain. I made sure to upsample the wav for prosody, but still getting rubbish. Could you point me where I am wrong ?
olup commented
I have been pointed to a bug in the notebook's code at this line
the start of the shell should be:
ps = global_phonemizer.phonemize([text])
# ps = word_tokenize(ps[0]) // this line is wrong
tokens = textclenaer(ps[0])
yl4579 commented
Sorry I made a mistake in the update. I have fixed it, can you try if it works now?
olup commented
Hey, thanks, it's effectively now working, and I'm closing my issue!
But while i'm at it, would you have pointers on to have better end of sentences ? Model produces no significant pause between sentences and tend to have the wrong intonation at the end.