compute_per_example_losses()

TensorSpeech / TensorFlowTTS

:stuck_out_tongue_closed_eyes: TensorFlowTTS: Real-Time State-of-the-art Speech Synthesis for Tensorflow 2 (supported including English, French, Korean, Chinese, German and Easy to adapt for other languages)

https://tensorspeech.github.io/TensorFlowTTS/

Apache License 2.0

3.81k stars 811 forks source link

compute_per_example_losses() #764

Closed MostafaAlaviyan closed 2 years ago

MostafaAlaviyan commented 2 years ago

hi thank you for the valuable repo. I trained fastspeech2 on multi-speaker datasets. When I add a new speaker, the Mel-spectra is very inaccurate. Now, I want to compute the loss to determine which sample(s) cases this issue. I see an attribute named "compute_per_example_losses" in "train_fastspeech2.py ". How can I use this function, could you please explain? What is the batch parameter? do you have an example? Best regards. @dathudeptrai @ZDisket

stale[bot] commented 2 years ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.