TextToSpeech device always automatically selected

neonbjb / tortoise-tts

A multi-voice TTS system trained with an emphasis on quality

Apache License 2.0

12.35k stars 1.73k forks source link

Open damithsenanayake opened 1 month ago

damithsenanayake commented 1 month ago

In a multi GPU setting, the GPU is always autmatically selected. This causes the voice presets to sometimes be loaded on a separate GPU.

Possible Fix. Include a check to see if the device is set in api.py/TextToSpeech

damithsenanayake commented 1 month ago

Possible fix added here.