search

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
3.84k stars 815 forks source link

Performance Benchmark #765

Closed iprovalo closed 2 years ago

iprovalo commented 2 years ago

Is there any data on performance benchmark for various GPU memory / batch size configurations? I am using 15gb single GPU. Unless I use 16 for batch size, it throws OOM. With 16 batch size it takes 36 hours to run 16k steps on en_US data for examples/tacotron2/train_tacotron2.py.

It took 21 hours to run 200k steps for examples/multiband_melgan/train_multiband_melgan.py

Are these expected?

Thank you!

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.