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

Merge pull request #1343 from mi804/ltx2.3_multiref

Browse Source 
Ltx2.3 multiref
This commit is contained in:
Hong Zhang
2026-03-10 17:31:05 +08:00
committed by GitHub
parent f3ebd6f714
commit c927062546
4 changed files with 113 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
diffsynth/diffusion/base_pipeline.py
View File

@@ -147,6 +147,19 @@ class BasePipeline(torch.nn.Module):
video = [self.vae_output_to_image(image, pattern="H W C", min_value=min_value, max_value=max_value) for image in vae_output]
return video
def output_audio_format_check(self, audio_output):
# output standard foramt: [C, T], output dtype: float()
# 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):
if self.vram_management_enabled: