Music and sound effects with Guaardvark
Read setup first; needs the audio_foundry plugin. B=${GUAARDVARK_URL:-http://localhost:5000}.
ACE-Step takes ~10 GB VRAM; the orchestrator evicts other models while it runs.
A song
curl -s -X POST $B/api/audio-foundry/generate/music -H 'Content-Type: application/json' -d '{
"style_prompt": "indie folk, fingerpicked acoustic guitar, warm male vocal, 96 bpm, intimate",
"lyrics": "[verse]\nLines here\n[chorus]\nHook here",
"negative_prompt": "distorted, muddy, synth",
"duration_s": 90, "instrumental_only": false,
"output_format": "wav", "seed": 1234, "async": true
}'
style_prompt uses ACE-Step's tag vocabulary: genre, instruments, mood, tempo, vocal type.
Vague words ("professional", "futuristic") drift to the model's prior; be concrete, and use
negative_prompt to push away from it. POST $B/api/audio-foundry/rewrite-music-prompt
turns plain English into the tag form when the user wants help.lyrics with [verse] / [chorus] / [bridge] markers; omit or set instrumental_only: true.duration_s up to 240. Always send "async": true for anything over a few seconds.- Response
202 {"job_id"} → poll GET $B/api/audio-foundry/jobs/<job_id>; done when status
is done, with path, document_id, duration_s, seed. Reuse the seed to reproduce a take.
A sound effect or ambience
curl -s -X POST $B/api/audio-foundry/generate/fx -H 'Content-Type: application/json' -d '{
"prompt": "rain on a tin roof, distant thunder, no music", "duration_s": 20, "output_format": "wav", "async": true
}'
duration_s up to 47. Same job polling.
Then
- A finished song can go straight into the music skill-video as the
song.
- Jobs:
GET $B/api/audio-foundry/jobs lists, DELETE clears finished ones.
- Attribution comes back in the job result; keep it with the file if the user publishes.
1---2name: music3description: Generate full songs with vocals or instrumentals (ACE-Step) and sound effects or ambience (Stable Audio Open) on the user's GPU through Guaardvark's Audio Foundry. Use when the user asks for a song, a beat, a jingle, background music, or a sound effect, made locally with no subscription.4---56# Music and sound effects with Guaardvark78Read `setup` first; needs the `audio_foundry` plugin. `B=${GUAARDVARK_URL:-http://localhost:5000}`.9ACE-Step takes ~10 GB VRAM; the orchestrator evicts other models while it runs.1011## A song1213```bash14curl -s -X POST $B/api/audio-foundry/generate/music -H 'Content-Type: application/json' -d '{15 "style_prompt": "indie folk, fingerpicked acoustic guitar, warm male vocal, 96 bpm, intimate",16 "lyrics": "[verse]\nLines here\n[chorus]\nHook here",17 "negative_prompt": "distorted, muddy, synth",18 "duration_s": 90, "instrumental_only": false,19 "output_format": "wav", "seed": 1234, "async": true20}'21```22- `style_prompt` uses ACE-Step's tag vocabulary: genre, instruments, mood, tempo, vocal type.23 Vague words ("professional", "futuristic") drift to the model's prior; be concrete, and use24 `negative_prompt` to push away from it. `POST $B/api/audio-foundry/rewrite-music-prompt`25 turns plain English into the tag form when the user wants help.26- `lyrics` with `[verse]` / `[chorus]` / `[bridge]` markers; omit or set `instrumental_only: true`.27- `duration_s` up to 240. Always send `"async": true` for anything over a few seconds.28- Response `202 {"job_id"}` → poll `GET $B/api/audio-foundry/jobs/<job_id>`; done when `status`29 is `done`, with `path`, `document_id`, `duration_s`, `seed`. Reuse the seed to reproduce a take.3031## A sound effect or ambience3233```bash34curl -s -X POST $B/api/audio-foundry/generate/fx -H 'Content-Type: application/json' -d '{35 "prompt": "rain on a tin roof, distant thunder, no music", "duration_s": 20, "output_format": "wav", "async": true36}'37```38`duration_s` up to 47. Same job polling.3940## Then4142- A finished song can go straight into the music skill-video as the `song`.43- Jobs: `GET $B/api/audio-foundry/jobs` lists, `DELETE` clears finished ones.44- Attribution comes back in the job result; keep it with the file if the user publishes.
guaardvark