search

huggingface / parler-tts

Inference and training library for high-quality TTS models.
Apache License 2.0
2.6k stars 265 forks source link

Unable to get it to run on the gpu. #29

Open Neamur opened 4 weeks ago

Neamur commented 4 weeks ago

Hey, I was trying to run the code on a virtual python env, and the tts doesn't seem to use the gpu on my system.

Do we need to have cuda toolkit installed for the gpu to be used?

Neamur commented 4 weeks ago

Also, since it is running on the cpu right now, would it be possible to limit the processing to half the cores rather than full core all threads workload? It taxes the Cpu quite a lot.

image

Would highly appreciate if we could limit how much cores/threads the audio generation uses.

ylacombe commented 1 week ago

Do we need to have cuda toolkit installed for the gpu to be used?

No toolkit that I'm aware of! ii should run on GPU automatically, can you send the bash command you're using?

Neamur commented 1 week ago

I was using the exact code snippet provided in the Readme. Could it be that due to less vram it defaults to cpu? I have 4 gb vram.