theluyuan/DiffSynth-Studio
1
0
Fork 0
mirror of https://github.com/modelscope/DiffSynth-Studio.git synced 2026-03-18 22:08:13 +00:00
Code Issues Packages Projects Releases Wiki Activity

Mova (#1337)

Browse Source 
* support mova inference

* mova media_io

* add unified audio_video api & fix bug of mono audio input for ltx

* support mova train

* mova docs

* fix bug
This commit is contained in:
Hong Zhang
2026-03-13 13:06:07 +08:00
committed by GitHub
parent 4741542523
commit 681df93a85
37 changed files with 3102 additions and 181 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
diffsynth/diffusion/base_pipeline.py
View File

@@ -152,13 +152,6 @@ class BasePipeline(torch.nn.Module):
# remove batch dim
if audio_output.ndim == 3:
audio_output = audio_output.squeeze(0)
# Transform to stereo
if audio_output.shape[0] == 1:
audio_output = audio_output.repeat(2, 1)
elif audio_output.shape[0] == 2:
pass
else:
raise ValueError("The output audio should be [C, T] or [1, C, T] or [2, C, T].")
return audio_output.float()
def load_models_to_device(self, model_names):