Performance Trade-off - Reduce speed for better memory usage in TTS #4040
Unanswered
cuneyttyler
asked this question in
General Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I'm running several TTS models. Currently I'm experimenting with your_tts model which is a smaller model. I have 4 GB memory GPU (GTX 1050) with Cuda support. For some time, it runs without problem and very fast. After a while, it throws memory error. I wonder if there's a way to reduce speed and use less memory? Through config.json or something like that, maybe in code through some parameters?
Beta Was this translation helpful? Give feedback.
All reactions