Slow Audio Generation – Output Voice Slower Than Input
#2
by
Vicky15
- opened
The generated audio from the PlayDiffusion model is significantly slower than expected. I tested it both locally and on the Hugging Face Space, and in both cases, the output voice duration is longer than the original input with the same transcript. This results in a slow, unnatural-sounding voice. Please let me know if this is a known issue or if there’s a way to adjust the speed.
Answered in github. Let's keep the discussion there - https://github.com/playht/PlayDiffusion/issues/11
yavorr
changed discussion status to
closed