ComfyUI Node
AceStep 1.5 SFT Get Music Infos
Analyzes audio with the native ACE-Step Transcriber to extract lyric, vocal and song-structure tags, plus BPM and key/scale via librosa. Wire outputs to Generate node or to text display nodes to inspect results.
AceStep 1.5 SFT Get Music Infos
- audio
- tags
- bpm
- keyscale
- music_infos
◄get_tagstrue►
◄get_bpmtrue►
◄get_keyscaletrue►
◄max_new_tokens256►
◄audio_duration60►
◄unload_modeltrue►
◄use_flash_attnfalse►
◄temperature0.00►
◄top_p1.00►
◄top_k0►
◄repetition_penalty1.10►
◄seed0►
Categoryaudio/AceStep SFT
Inputs (13)
| Name | Type | Default | Description |
|---|---|---|---|
| audio | AUDIO | Audio to analyze for style, BPM and key/scale. | |
| get_tags | BOOLEAN | true | Extract descriptive tags from the audio using the native ACE-Step transcriber. |
| get_bpm | BOOLEAN | true | Detect BPM from audio using librosa. |
| get_keyscale | BOOLEAN | true | Detect key and scale from audio using librosa. |
| max_new_tokensopt | INT | 25664–2000 | Maximum tokens for the native ACE-Step transcription output. Higher values preserve more lyric/structure detail. |
| audio_durationopt | INT | 6010–300 | Max seconds of audio to analyze (center crop). ACE-Step Transcriber benefits from more context for sections and lyrics. |
| unload_modelopt | BOOLEAN | true | Unload the ACE-Step transcriber after use to free VRAM for generation. |
| use_flash_attnopt | BOOLEAN | false | Use FlashAttention-2 for the ACE-Step transcriber. Requires flash-attn package installed. Faster and uses less VRAM. |
| temperatureopt | FLOAT | 0.000–2 | Sampling temperature for transcription generation. 0 = deterministic and recommended for stable structure extraction. |
| top_popt | FLOAT | 1.000–1 | Nucleus sampling for transcription generation. Keep at 1.0 for native deterministic behavior unless you are experimenting. |
| top_kopt | INT | 00–200 | Top-K sampling for transcription generation. 0 preserves the native default behavior. |
| repetition_penaltyopt | FLOAT | 1.101–3 | Penalty against repeated tokens in the transcription output. 1.1 is a mild safeguard against loops. |
| seedopt | INT | 00–18446744073709550000 | Random seed for reproducible transcription generation when sampling is enabled. |
Outputs (4)
| Name | Type | Description |
|---|---|---|
| tags | STRING | — |
| bpm | INT | — |
| keyscale | STRING | — |
| music_infos | STRING | — |