You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi I have some question about the silence label. I train an acoustic model from scratch, and I find that the alignment result doesn't contain the 'sil' label that represents the silence. Even after I add the two line "{LG} spn" and " {SL} sil" into the dictionary as the documentation said, the 'sil' label is still missed. In the past, when I use the pretrianed acoustic model, the 'sil' and 'spn' symbol will appear. May I know how to fix this problem? Thank you a lot!
The text was updated successfully, but these errors were encountered:
So I recently removed the outputting of "sil" and "sp" to simplify the export logic and clean up the output textgrids. Do you need these outputted? You should be able to just use any blank output in the textgrid to mean silence.
"spn" should still be output correctly (i.e. for unknown words), but I haven't tested the "{LG}" case in a long while, I'll try to take a look at it to verify it.
Thank you for your reply! I am glad to know this is not my implementation problem but a new property of the new MFA version. I was stucked that I can't access the interval with '' label. But I find this can be fixed by upgrading the version of praatio.
Hi I have some question about the silence label. I train an acoustic model from scratch, and I find that the alignment result doesn't contain the 'sil' label that represents the silence. Even after I add the two line "{LG} spn" and " {SL} sil" into the dictionary as the documentation said, the 'sil' label is still missed. In the past, when I use the pretrianed acoustic model, the 'sil' and 'spn' symbol will appear. May I know how to fix this problem? Thank you a lot!
The text was updated successfully, but these errors were encountered: