elevenlabs_act
Generate audio with ElevenLabs. EVERY call spends the account's credits
(characters for speech, seconds for music and sound) and is not retried:
read elevenlabs op='get_subscription' before a batch. On the shared key, text
is at most 5,000 characters per call (text_too_long): cut a longer script
into lines, or use your own key, which lifts the limit.
On the SHARED key (granted by oto, no key of your own), only the generations
this instance serves run — by default create_speech; the others ask for your
own ElevenLabs key.
The audio comes back as a signed URL (url, valid expires_in seconds,
deleted after a day), never inline: download it right away.
op:
create_speech—textread byvoice_id(fromelevenlabsop='list_voices'), withmodel_id,voice_settings,seedandlanguage_codeoptional.previous_text/next_textmake a line sound continuous with its neighbours when a script is read line by line. Withtimestamps=truethe answer also carrieswords(each word's start and end, seconds) and the raw per-characteralignment.create_music— a track from aprompt(lengthmusic_length_ms, 3,000 to 300,000) or from acomposition_plan(sections, each with its duration and positive and negative styles), never both;respect_sections_durations,force_instrumental,seedoptional. A prompt naming an artist or a copyrighted work is refused. Billed per minute of track (~900 credits a minute). On the shared key, when the instance serves it: the length is required (music_length_ms, or each section'sduration_ms), the quota is checked against it.create_sound— a sound effect described intext;duration_seconds(0.5 to 30, else the model decides),prompt_influence(0 to 1),loopoptional.create_transcript— the words said in an audio file (Scribe):sourcedesignates the file (a{"kind": "transit", "ref"}an oto tool returned, a project file, a URL…),language_codepins the language (else detected),audio_secondsits length. Returnstext,language_codeandwords(each with its start and end, seconds). Billed per second of audio (~330 credits a minute). On the shared key, when the instance serves it:audio_secondsis required, at most 300 seconds and 1 MB of audio per call.create_voice_design— up to three previews of a voice described invoice_description, each readingtext(or a text the model writes withauto_generate_text=true); each preview has itsgenerated_voice_idand its own URL. Saving a preview as a voice of the account is not done here.
output_format is <codec>_<sample rate>[_<bitrate>], default
mp3_44100_128; some (mp3_44100_192, pcm_44100) are paid plans only.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"op": "<op>"
},
"name": "elevenlabs_act"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| op | Yes | What to generate (see above). | |
| _org | No | Org (id) de CET APPEL. `oto_list_orgs()`. | |
| loop | No | Sound only: make it loop seamlessly. | |
| seed | No | 0 to 4294967295, to make a take reproducible as far as the model allows. | |
| text | No | The words to read (`create_speech`), the sound to make (`create_sound`), or the preview's line (`create_voice_design`); at most 5000 characters on the shared key. | |
| _group | No | Équipe (id) de CET APPEL, sous son org parente. | |
| prompt | No | Music only: the track described in words. | |
| source | No | Transcription only: the audio file, e.g. {"kind": "transit", "ref": "..."} or {"kind": "url", "url": "https://..."}. | |
| _run_id | No | OBLIGATOIRE à chaque appel dès `run_start` ou `data_claim_next` : sans lui, l'écriture sur une ligne réservée est refusée. Hérité seulement si le serveur tient un run actif — n'y compte pas. | |
| _project | No | Projet (id) de CET APPEL — jeton PRIMAIRE. `oto_project(op='list')`. | |
| model_id | No | Model id, e.g. eleven_multilingual_v2 or eleven_v3 for speech, music_v1 for music; the API's default when omitted. | |
| voice_id | No | The voice that reads, for `create_speech`. | |
| _instance | No | Instance de connecteur PRÉCISE (un credential exact). `oto_instance(op='list')`. | |
| next_text | No | Speech only: the line after this one, for continuity. | |
| timestamps | No | Speech only: also return each word's start and end. | |
| audio_seconds | No | Transcription only: the clip's length in seconds — what the quota is checked against before the call; required on the shared key. | |
| language_code | No | Speech and transcription: ISO 639-1 code to force the language, when the model supports it. | |
| output_format | No | e.g. mp3_44100_128 (default), mp3_48000_192, pcm_24000, opus_48000_64. | |
| previous_text | No | Speech only: the line before this one, for continuity. | |
| voice_settings | No | Speech only: {stability, similarity_boost, style, use_speaker_boost, speed}, each optional. | |
| music_length_ms | No | Music from a prompt only: the length, 3000 to 300000. | |
| composition_plan | No | Music only: {positive_global_styles, negative_global_styles, sections: [{section_name, duration_ms, positive_local_styles, negative_local_styles, lines}]}. | |
| duration_seconds | No | Sound only: 0.5 to 30 seconds. | |
| prompt_influence | No | Sound only: how literally to follow the text, 0 to 1. | |
| voice_description | No | Voice design only: the voice in words, e.g. age, accent, tone, pace. | |
| auto_generate_text | No | Voice design only: let the model write the preview's line. | |
| force_instrumental | No | Music only: no vocals. | |
| respect_sections_durations | No | Music from a plan only: keep each section's duration exactly. |