AI/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-27 17:22:53 +03:00
Code Activity

[AudioLDM] Fix dtype of returned waveform (#3189)

Browse Source
This commit is contained in:
Sanchit Gandhi
2023-04-21 18:24:37 +00:00
committed by GitHub
parent 11f527ac0f
commit 90eac14f72
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/diffusers/pipelines/audioldm/pipeline_audioldm.py
View File

@@ -293,7 +293,7 @@ class AudioLDMPipeline(DiffusionPipeline):
waveform = self.vocoder(mel_spectrogram)
# we always cast to float32 as this does not cause significant overhead and is compatible with bfloat16
waveform = waveform.cpu()
waveform = waveform.cpu().float()
return waveform
# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_extra_step_kwargs