MontrealCorpusTools / Montreal-Forced-Aligner

Command line utility for forced alignment using Kaldi
https://montrealcorpustools.github.io/Montreal-Forced-Aligner/
MIT License
1.31k stars 244 forks source link

Is it possible to only run MFA on a subset of speaker folders? #836

Closed jasonppy closed 3 days ago

jasonppy commented 3 days ago

Thanks for the great tool!

I have a folder called A, under which there are >10000 speakers folders, and if I just run mfa align on folder A, it will be very slow, and might OOM. Therefore I wonder if I can parallize it by run mfa align on a subset of speaker folders say 100 folders in each job.

Thanks!

mmcauliffe commented 3 days ago

You can increase the number of parallel jobs via --num_jobs: https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/alignment.html#cmdoption-mfa-align-j, the default is 3 so increasing that will speed it up.