search

qiuqiao / SOFA

SOFA: Singing-Oriented Forced Aligner
MIT License
116 stars 17 forks source link

can't train model because ph_seq is set to None #21

Closed vocatart closed 4 months ago

vocatart commented 4 months ago

I don't know enough about python to be able to fix this problem with a pr, so I hope this is descriptive enough

commit 7cd5771 adds wav_length to _infer_once, but doesn't change arguments provided in validation_step https://github.com/qiuqiao/SOFA/blob/7cd57710a31b7d7b9ca760adaa8fa2dc6c9ff15e/modules/task/forced_alignment.py#L774-L781 this sets ph_seq to None, causing _infer_once to fail with a NoneType error when using train.py. I reverted back to commit 2591e98 and this error was resolved.

qiuqiao commented 4 months ago

I'm sorry for submitting the commit without adequate testing. I will be more careful in the future.

vocatart commented 4 months ago

don't worry about it! I'm really happy with the results I'm getting with SOFA so far! keep up the good work :)