eleven_v3

The expressive generation, and the only ElevenLabs model that supports OnCore's emotion-tagging delivery mode on English calls. It exposes stability alone. Audio tags are on Eleven v3 audio tags.

Reached through the ElevenLabs API with OnCore's own API key, not the LiveKit Inference gateway.

Voice settings are set on the assistant, not in the voice body. Leave every one unset and ElevenLabs' own defaults apply; set any one of them and OnCore sends stability and similarityBoost too, at the same values ElevenLabs defaults to.

Settings#

Every setting is optional. Leave one out and OnCore sends nothing for it, so the Default applies. If invalid is what the provider does with a value it does not accept.

Overriding on a single call#

Return this from your assistant-request webhook to change these settings for one call only. The stored voice is left untouched, and the next call starts from it again.

{
  "data": {
    "overrides": {
      "voice": {
        "settings": {
          "stability": 0.5
        }
      }
    }
  }
}

Send only the keys you want to change: anything you leave out keeps the value the assistant already has. Add "id" beside "settings" to swap the whole voice for another one from your catalogue at the same time.

ElevenLabs presents v3 stability as three named settings rather than a slider, and they map onto the numbers you send: 0.0 is Creative (expressive and varied, and it can even sing), 0.5 is Natural (conversational balance), 1.0 is Robust (precise, stable and predictable). The API accepts any value between 0 and 1, but those three are the ones the model is tuned around, so prefer them over a value in between.

ElevenLabs exposes stability alone for v3. similarityBoost, speed, style and useSpeakerBoost are accepted by the API without error, but they are not part of this model's controls, so do not expect them to change anything.

For the tag catalogue, the closed set emotion tagging injects, and how to write [tag] yourself, see Eleven v3 audio tags.

Next steps#

• eleven_v4_turbo: the next ElevenLabs model.
• eleven_multilingual_v2: the previous ElevenLabs model.
• Providers introduction: per-call overrides, and how a wrong value behaves across providers.