Open dan-ya opened 7 months ago
Can you try running conda update -c conda-forge kalpy
and retry?
Thank you! Now it is fine.
There is another strange issue. It got stuck at the Exporting alignment TextGrids to ...
stage:
INFO Setting up corpus information...
INFO Found 6 speakers across 123452 files, average
number of utterances per speaker:
20575.333333333332
INFO Jobs already initialized.
INFO Text already normalized.
INFO Generating MFCCs...
100% ━━━━━━━━━━ 123,452/1… [ 0:03:41 < 0:00:00 , 384 it/s ]
INFO Calculating CMVN...
INFO Generating final features...
100% ━━━━━━━━━ 123,452/… [ 0:00:09 < 0:00:00 , 14,900 ]
it/s
INFO Creating corpus split...
100% ━━━━━━━━━ 123,452/… [ 0:00:04 < 0:00:00 , 40,003 ]
it/s
INFO Compiling training graphs...
INFO Performing first-pass alignment...
INFO Generating alignments...
100% ━━━━━━━━━━ 123,452/1… [ 0:05:15 < 0:00:00 , 250 it/s ]
INFO Calculating fMLLR for speaker adaptation...
99% ━━━━━━━━╸ 121,921/1… [ 0:01:12 < 0:00:02 , 1,451 ]
it/s
INFO Performing second-pass alignment...
INFO Generating alignments...
100% ━━━━━━━━━━ 123,452/1… [ 0:04:19 < 0:00:00 , 300 it/s ]
INFO Collecting phone and word alignments from
alignment lattices...
99% ━━━━━━━━╸ 122,119/1… [ 0:00:21 < 0:00:01 , 6,662 ]
it/s
WARNING Alignment analysis not available without using
postgresql
INFO Exporting alignment TextGrids to
Providence_mfa_aligned.arpa...
0% ━━━━━━━━━━━━ 0/123,452 [ 1:04:02 < -:--:-- , ? it/s ]
Debugging checklist
[x] Have you read the troubleshooting page (https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/troubleshooting.html) and searched the documentation to ensure that your issue is not addressed there? [x] Have you updated to latest MFA version (check https://montreal-forced-aligner.readthedocs.io/en/latest/changelog/changelog_3.0.html)? What is the output of
mfa version
? [x] Have you tried rerunning the command with the--clean
flag?Describe the issue I want to align some data from Providence Corpus (with .lab files as input). The same error happens both with
--clean
flag and without. I have tried to validate the corpus with a following command and got the same error:mfa validate archive/Providence_mfa_dataset english_us_arpa english_us_arpa --clean
Both the model and the dictionary are pre-trained and were just downloaded with:
The output is the following:
Desktop (please complete the following information):