Generate AI audio
generate_audioStarts an audio generation on the signed-in aicut account (speech, music, a song with vocals or a sound effect, chosen by model) and returns its job id at once; it does NOT wait for the audio. WHEN: the user asks for one of those. Call list_models with type: "audio" first: each model takes DIFFERENT arguments (its parameters; get_recipe starter-audio maps them), and a field it does not publish is rejected, not ignored. A speech voice id comes from list_voices. HOW IT SOUNDS: the knobs marked 'Result only' come at no extra price and need no re-quote; only those a model publishes in parameters are accepted. THE CARD: this call already opened the aicut card: do not call show_generation for this job. WAIT: wait_for_generation (this id, media), again at once while terminal: false, then one short line; no get_* polling, timers, background shells or narrating the wait. No card? Give the user the url. TIMING: audio usually finishes in under a minute (often 10 seconds for a short effect) and can take up to about 4 minutes for a long script. Do not spawn timers or background shells. COST: speech, music and sound effects are priced on a RATE (per character, minute, second: pricing_rates), a song per call. Never multiply it yourself: pass estimate_only: true with the exact settings whenever the user has not agreed to the cost. SPENDS TOKENS: price with estimate_only: true, ask in the SAME message as the price, and wait for a go that names THIS purchase (SPENDING RULES, this server's instructions). IDEMPOTENCY: a reused idempotency_key REPLAYS the first job whatever the settings: never reuse one for a new generation (this server's instructions).
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| loop | No | Sound effects: loop seamlessly. Result only, not price. | |
| text | No | Speech: the exact words to speak (priced PER CHARACTER: never padded or shortened to change the price). Sound effect: a description of the sound. Only if the user asked; never to save tokens, never carried over. | |
| model | Yes | Model id exactly as `list_models` returns it (`elevenlabs-tts-v2` speech, `elevenlabs-music` music, `suno-v6` song, `elevenlabs-sfx` sound effect); it decides which other arguments apply. Use the one the user asked for. | |
| speed | No | Speech, `elevenlabs-tts-v2` only: speaking rate, 1 is normal. Result only. | |
| style | No | Speech, `elevenlabs-tts-v2` only: how far the speaker's own style is exaggerated. Result only. | |
| title | No | Song: the title (required in Lyrics mode). | |
| voice | No | Required by speech models. Get it from `list_voices` with `provider: 'elevenlabs', usable_for: 'audio', limit: 100` (21 stock voices; without `limit` you see 20 and cannot tell). The user picks it from `preview_url` links BEFORE this call. Never invent one: it fails AFTER the job exists and is charged. Cloned and custom voices work on `generate_fake_text_video` and `generate_image_story`, not here. | |
| lyrics | No | Song, Lyrics mode: words to sing; omit to have them written. | |
| prompt | No | Music: the track (mood, instruments, tempo). Song: the Quick-mode idea. | |
| duration | No | Sound-effect length in seconds within the model's `min`/`max`; billed per second. An omitted length is quoted at a fallback price. Only if the user asked; never to save tokens, never carried over. | |
| length_ms | No | Music length in whole milliseconds within the model's `min`/`max`; billed per started minute. An omitted length is quoted at a fallback price. Only if the user asked; never to save tokens, never carried over. | |
| stability | No | Speech: how steady the delivery is (lower more expressive, higher flatter). Result only, not price. | |
| custom_mode | No | Song: true = Lyrics mode (needs `title`; takes `lyrics`, `vocal_gender`, `duration_seconds`); omit = Quick (needs `prompt`). | |
| music_style | No | Song, required: genre, mood, instruments. | |
| vocal_gender | No | Song, Lyrics mode: singer voice (a preference). | |
| estimate_only | No | Price only; creates nothing. The only accurate audio quote. | |
| idempotency_key | No | Retry-safety key (IDEMPOTENCY). | |
| duration_seconds | No | Song, Lyrics mode: whole seconds. Result only. | |
| prompt_influence | No | Sound effects: how literally to follow the text (the model's `min`/`max`). Result only, not price. | |
| similarity_boost | No | Speech, `elevenlabs-tts-v2` and `-v4` only: how closely the delivery sticks to the voice. Result only. | |
| force_instrumental | No | Music: no vocals. Result only, not price. |