search

Stability-AI / stable-audio-tools

Generative models for conditional audio generation
MIT License
2.72k stars 258 forks source link

does it support standard v-diffusion? #133

Closed GenjiB closed 1 month ago

GenjiB commented 3 months ago

I notice this line: https://github.com/Stability-AI/stable-audio-tools/blob/main/stable_audio_tools/inference/generation.py#L237

Does the pretrained model support this v-diffusion?

zqevans commented 1 month ago

It does, even the k-diffusion calls are actually wrapped in a VDenoiser so they're technically running v-diffusion as well. The sample function in sampling.py is v-diffusion DDIM.