Hello, I've done training and decoding. I'm curious whether Is it possible to see the scores in decoding using monophones and triphones? If yes, where can I view it?
Thanks a lot.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It depends on what scores do you want to see - frame scores, segment scores, etc. Also, cmusphinx models use context dependent senones, not triphones, see http://cmusphinx.github.io/wiki/tutorialconcepts.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello, I've done training and decoding. I'm curious whether Is it possible to see the scores in decoding using monophones and triphones? If yes, where can I view it?
Thanks a lot.
It depends on what scores do you want to see - frame scores, segment scores, etc. Also, cmusphinx models use context dependent senones, not triphones, see http://cmusphinx.github.io/wiki/tutorialconcepts.
Thank you for your response Sir Nickolay. Does sphinx have scoring like kaldi in the WERs in using monophones, etc?
Thanks a lot!
I'm not sure what exact feature are you talking about. The sphinxtrain training process reports the word error rate.
Hi Nickolay. Sorry I'm not able to reply this past days. Anyway, does pocketsphinx or sphinxtrain train with monophones and triphones? Thanks!