Open thangdc94 opened 4 years ago
I think there may be an 'issue' on github about this, and maybe even a PR; see if you can find it.
On Tue, Oct 20, 2020 at 10:44 AM Phạm Ngọc Thắng notifications@github.com wrote:
I'm using compute-gop to get gop value. I expected number of phones in gop file should be equal to number of phones in alignment. But there are some exceptions when same phones are placed next to each other:
E.g: bookcase B UH1 K K EY2 S There are 2 K And I got the following output: phone.1.ctm
utt 1 0.000 0.020 1 utt 1 0.020 0.010 107 # B utt 1 0.030 0.040 305 # UH1 utt 1 0.070 0.030 221 # K utt 1 0.100 0.050 221 # K utt 1 0.150 0.050 169 # EY2 utt 1 0.200 0.070 280 # S utt 1 0.270 0.090 1
gop.1.txt
utt [ 1 -0.5976977 ] [ 107 -0.193491 ] [ 305 -1.435675 ] [ 221 -0.8830113 ] [ 169 -1.103366 ] [ 280 -2.453551 ] [ 1 -5.452305 ]
K is missing in gop.1.txt
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/kaldi-asr/kaldi/issues/4302, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAZFLO2MUESUVJJA67XOURTSLT2PNANCNFSM4SXKBBBQ .
This issue has been automatically marked as stale by a bot solely because it has not had recent activity. Please add any comment (simply 'ping' is enough) to prevent the issue from being closed for 60 more days if you believe it should be kept open.
I'm using compute-gop to get gop value. I expected number of phones in gop file should be equal to number of phones in alignment. But there are some exceptions when same phones are placed next to each other:
E.g:
bookcase B UH1 K K EY2 S
There are 2 K (phone id = 221) And I got the following output:phone.1.ctm
gop.1.txt
K is missing in gop.1.txt