mrleo1nid/ACE-Step
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Compare commits

base: mrleo1nid/ACE-Step:6ae0852b1388de6dc0cca26b31a86d711f723cb3
Branches Tags
mrleo1nid/ACE-Step:main
..
compare: mrleo1nid/ACE-Step:1bee4c9f5b43e30995f8d4d33b3919197ce1bd68
Branches Tags
mrleo1nid/ACE-Step:main
5 Commits
6ae0852b13 .. 1bee4c9f5b
Author SHA1 Message Date
Gong Junmin 1bee4c9f5b Merge pull request #373 from iackov/fix/extend-mode-shape-mismatch 2026-02-15 12:57:54 +08:00
Gong Junmin 4a24cfb662 Merge pull request #375 from ace-step/copilot/add-external-link-for-model 2026-02-15 12:57:18 +08:00
copilot-swe-agent[bot]andChuxiJ 66b690b05b Add external link for ACE-Step v1.5 model 
Co-authored-by: ChuxiJ <30956809+ChuxiJ@users.noreply.github.com>
 2026-01-28 00:09:54 +00:00
copilot-swe-agent[bot] bf3510b949 Initial plan 2026-01-28 00:08:27 +00:00
jackj 435e9fd667 Fix: Shape mismatch in extend mode causing AssertionError 
- Added automatic shape alignment for target_latents and x0
- Handles both shorter (padding) and longer (trimming) cases
- Fixes crash in extend mode with long audio files
- Minimal impact on audio quality (~0.05-0.15 sec)

Resolves issue where extend mode fails with AssertionError
when target_latents shape doesn't match x0 shape after
padding/trimming operations.
 2026-01-23 23:15:42 +05:00
2 changed files with 17 additions and 1 deletions
Expand all files Collapse all files
README.md
+4 -1
View File
@@ -6,7 +6,8 @@
<a href="https://modelscope.cn/models/ACE-Step/ACE-Step-v1-3.5B">ModelScope</a> |
<a href="https://huggingface.co/spaces/ACE-Step/ACE-Step">Space Demo</a> |
<a href="https://discord.gg/PeWDxrkdj7">Discord</a> |
<a href="https://arxiv.org/abs/2506.00045">Technical Report</a>
<a href="https://arxiv.org/abs/2506.00045">Technical Report</a> |
<a href="https://ace-step.github.io/ace-step-v1.5.github.io/">ACE-Step v1.5</a>
</p>
<p align="center">
@@ -32,6 +33,8 @@ Rather than building yet another end-to-end text-to-music pipeline, our vision i
## 📢 News and Updates
- 🎉 **2026.01.28:** Released [ACE-Step v1.5](https://ace-step.github.io/ace-step-v1.5.github.io/) - Our latest and most advanced model is now available!
- 📃 2025.06.02: Released [ACE-Step Technical Report (PDF)](https://arxiv.org/abs/2506.00045).
- 🎮 2025.05.14: Add `Stable Audio Open Small` sampler `pingpong`. Use SDE to achieve better music consistency and quality, including lyric alignment and style alignment. Use a better method to re-implement `Audio2Audio`
acestep/pipeline_ace_step.py
+13
View File
@@ -1046,6 +1046,19 @@ class ACEStepPipeline:
if right_pad_frame_length > 0:
padd_list.append(retake_latents[:, :, :, -right_pad_frame_length:])
target_latents = torch.cat(padd_list, dim=-1)
# Fix shape mismatch between target_latents and x0
if target_latents.shape[-1] != x0.shape[-1]:
if target_latents.shape[-1] < x0.shape[-1]:
# Pad with zeros if target_latents is shorter
padding = x0.shape[-1] - target_latents.shape[-1]
target_latents = torch.nn.functional.pad(
target_latents, (0, padding), "constant", 0
)
else:
# Trim if target_latents is longer
target_latents = target_latents[..., :x0.shape[-1]]
assert (
target_latents.shape[-1] == x0.shape[-1]
), f"{target_latents.shape=} {x0.shape=}"