Spaces:

WeReCooking
/

ACE-Step-CPU

Running

App Files Files Community

ACE-Step-CPU

Commit History

May 1

03:07
compact UI: remove instrumental checkbox, lock steps, scroll fix, timing labels, LoRA+LM same row Running

Nekochu

20382cb
00:33
default 200 epochs, label with timing info, instant cancel in preprocessing

Nekochu

9151fbf

Apr 30

23:02
instant cancel during preprocessing (dont wait for thread to finish)

Nekochu

059d153
22:43
fix: use librosa instead of torchaudio for VAD (torchcodec not installed), fix audios deprecation

Nekochu

17d39ba
22:30
upgrade torch>=2.6 (CLAP needs torch.load fix for CVE-2025-32434)

Nekochu

6b3e616
21:55
fix: parse multipart/mixed response from /understand (was expecting JSON, got multipart)

Nekochu

9c293be
21:10
fix LM caption: use poll response data directly, empty body from /job?result=1 was the bug

Nekochu

92f884a
20:34
debug LM caption: log result keys, check nested result, increase fetch timeout to 120s

Nekochu

3590f35
19:18
fix: 5h fixed timeout per file, check total feasibility first

Nekochu

9d04583
19:14
LM captioning: 5h timeout per file, check feasibility before starting

Nekochu

5dedf2e
18:15
zip download: LoRA adapter + generated captions bundled together

Nekochu

89af747
17:38
remove lightning dep (not needed, breaks CPU torch index)

Nekochu

5d846d9
17:37
wire fast captioning (CLAP+Whisper+VAD) into training, add LM caption checkbox

Nekochu

d6a3e45
17:34
add fast captioning module (CLAP + faster-whisper + Silero VAD), update deps

Nekochu

4619f39
15:14
random 60s crop at training time (matches Side-Step chunk-duration), remove pre-split chunking

Nekochu

d3618ec
00:31
audio-level chunking (not latent), auto-scale epochs for chunk count

Nekochu

1ee8f1f

Apr 28

05:45
chunk latents into ~30s segments for faster CPU training, energy-aware boundaries

Nekochu

2e395ab
04:46
default rank 16 for faster CPU training

Nekochu

81f54b1
04:06
add credit lines to both tabs

Nekochu

afe7cf4
03:38
compact Generate Music UI, same layout style as Train tab

Nekochu

4d42fae
03:07
move audio upload to left column

Nekochu

8d2b494

Apr 27

12:35
skip bare librosa sidecar, let preprocessing faf analysis handle caption fallback

Nekochu

53f6566
12:33
fix adapter save path, smart LM fallback, compact training UI, remove Server Status

Nekochu

35fbf3e

Apr 26

15:46
cancel, captioning, preprocessing, sidecar upload, elapsed time, GeneratorExit fix

Nekochu

32de701
04:17
fix review: debug leak, int crash, rank mismatch, 0-byte skip, log cap, understand diag

Nekochu

4d9a556
03:52
fix: save PEFT adapter (not full model), remove random suffix from LoRA names, fix epoch cap to 1000

Nekochu

57df0f6
02:04
remove XL checkpoint download (OOMKilled build, training uses standard turbo)

Nekochu

6d9fb39
01:56
fix: save_every_n_epochs=0, add demucs-infer to Dockerfile, debug adapter dir

Nekochu

0e27e49

Apr 25

08:31
fix all review issues: dedup sampling/unwrap, thread-safe lock, cleanup, retry, security docs

Nekochu

829ed0c
08:19
update README with final state, full pipeline inference, LM generation step

Nekochu

a5741b1
08:12
fix inference: add LM generation step, detokenize codes before DiT, full pipeline working

Nekochu

ff9f4ad
07:08
add _is_space flag, block inference during training, understand clone fix

Nekochu

3c15b8b
06:40
fix understand_audio: clone tensors for inference mode, working on GPU (52s)

Nekochu

4b2f4ad
06:37
add understand_audio (LM reverse), demucs-infer fix, commit refs, dtype fixes

Nekochu

6bfdc38
05:14
major update: PyTorch inference, Gradio 6, session isolation, /understand captioning⁠

Nekochu

ff239f5
05:01
truncate long files to fit cap, show which files truncated/skipped

Nekochu

bc97006
04:59
accept files until total audio cap reached, skip rest with warning

Nekochu

956dc8c
04:58
replace per-file cap with total audio cap (30 min max)

Nekochu

1549b91
00:25
add LoRA download button after training (gr.File output, like rvc-beatrice)

Nekochu

2d3c27c

Apr 24

23:36
remove ace-server understand proxy, captioning stays librosa + txt sidecars

Nekochu

5b7a56f
23:28
add /understand API endpoint for GGUF audio captioning

Nekochu

a4457c3
23:24
SDPA first on Blackwell, FA2 only for Ampere/Hopper, txt caption support

Nekochu

04ccf32
23:17
add GPU/CUDA auto-detect, mixed precision, flash_attn, txt caption parser

Nekochu

917e4ed
23:05
update defaults: LR 3e-4, rank 32, alpha 2x rank (per Side-Step author)

Nekochu

04c031f
20:42
save only on cancel/finish, max epochs 1000, default 3

Nekochu

d42aa91
19:46
add mid/sas analysis modes (Demucs + ensemble), auto-select by dataset size

Nekochu

b38d0b1
19:33
add auto-captioning (BPM/key/signature via librosa), add librosa+mutagen deps

Nekochu

1d42836
19:24
switch training to standard turbo (11s/epoch), auto-select standard GGUF for LoRA inference

Nekochu

c0f2a13
17:55
add XL variant mappings to _VARIANT_DIR

Nekochu

3c5ce9c
08:22
fix: train on XL turbo (matches XL GGUF for inference), add XL checkpoint download

Nekochu

372f08e

Commit History

May 1

compact UI: remove instrumental checkbox, lock steps, scroll fix, timing labels, LoRA+LM same row Running

default 200 epochs, label with timing info, instant cancel in preprocessing

Apr 30

instant cancel during preprocessing (dont wait for thread to finish)

fix: use librosa instead of torchaudio for VAD (torchcodec not installed), fix audios deprecation

upgrade torch>=2.6 (CLAP needs torch.load fix for CVE-2025-32434)

fix: parse multipart/mixed response from /understand (was expecting JSON, got multipart)

fix LM caption: use poll response data directly, empty body from /job?result=1 was the bug

debug LM caption: log result keys, check nested result, increase fetch timeout to 120s

fix: 5h fixed timeout per file, check total feasibility first

LM captioning: 5h timeout per file, check feasibility before starting

zip download: LoRA adapter + generated captions bundled together

remove lightning dep (not needed, breaks CPU torch index)

wire fast captioning (CLAP+Whisper+VAD) into training, add LM caption checkbox

add fast captioning module (CLAP + faster-whisper + Silero VAD), update deps

random 60s crop at training time (matches Side-Step chunk-duration), remove pre-split chunking

audio-level chunking (not latent), auto-scale epochs for chunk count

Apr 28

chunk latents into ~30s segments for faster CPU training, energy-aware boundaries

default rank 16 for faster CPU training

add credit lines to both tabs

compact Generate Music UI, same layout style as Train tab

move audio upload to left column

Apr 27

skip bare librosa sidecar, let preprocessing faf analysis handle caption fallback

fix adapter save path, smart LM fallback, compact training UI, remove Server Status

Apr 26

cancel, captioning, preprocessing, sidecar upload, elapsed time, GeneratorExit fix

fix review: debug leak, int crash, rank mismatch, 0-byte skip, log cap, understand diag

fix: save PEFT adapter (not full model), remove random suffix from LoRA names, fix epoch cap to 1000

remove XL checkpoint download (OOMKilled build, training uses standard turbo)

fix: save_every_n_epochs=0, add demucs-infer to Dockerfile, debug adapter dir

Apr 25

fix all review issues: dedup sampling/unwrap, thread-safe lock, cleanup, retry, security docs

update README with final state, full pipeline inference, LM generation step

fix inference: add LM generation step, detokenize codes before DiT, full pipeline working

add _is_space flag, block inference during training, understand clone fix

fix understand_audio: clone tensors for inference mode, working on GPU (52s)

add understand_audio (LM reverse), demucs-infer fix, commit refs, dtype fixes

major update: PyTorch inference, Gradio 6, session isolation, /understand captioning⁠

truncate long files to fit cap, show which files truncated/skipped

accept files until total audio cap reached, skip rest with warning

replace per-file cap with total audio cap (30 min max)

add LoRA download button after training (gr.File output, like rvc-beatrice)

Apr 24

remove ace-server understand proxy, captioning stays librosa + txt sidecars

add /understand API endpoint for GGUF audio captioning

SDPA first on Blackwell, FA2 only for Ampere/Hopper, txt caption support

add GPU/CUDA auto-detect, mixed precision, flash_attn, txt caption parser

update defaults: LR 3e-4, rank 32, alpha 2x rank (per Side-Step author)

save only on cancel/finish, max epochs 1000, default 3

add mid/sas analysis modes (Demucs + ensemble), auto-select by dataset size

add auto-captioning (BPM/key/signature via librosa), add librosa+mutagen deps

switch training to standard turbo (11s/epoch), auto-select standard GGUF for LoRA inference

add XL variant mappings to _VARIANT_DIR

fix: train on XL turbo (matches XL GGUF for inference), add XL checkpoint download