Search the documentation

Pick a page and open it with Enter.

Speech Models

The speech models you can pick.

GET/api/v1/tts/models

The speech models the read_aloud tool can use. Filtered to what can run: no ELEVENLABS_API_KEY and only the free fallback remains, with the default shifting to it. No parameters.

Response

json
{
  "count": 5,
  "default": "eleven_multilingual_v2",
  "default_voice": "DDpANZ8PLYsm2RvgHVlV",
  "groups": ["ElevenLabs", "Free"],
  "models": [
    {
      "id": "eleven_multilingual_v2",
      "name": "Eleven Multilingual v2",
      "description": "The default — natural in many languages, understands IPA.",
      "group": "ElevenLabs",
      "runtime": "elevenlabs"
    },
    {
      "id": "free-google",
      "name": "Free voice",
      "description": "No key needed — a plain fallback voice.",
      "group": "Free",
      "runtime": "free"
    }
  ]
}
defaultstring

The model read_aloud uses when the request names none.

default_voicestring

The ElevenLabs voice used when the request sets no voice_id.

models[].runtimestring

elevenlabs or free — whether a key stands behind it.

200

List retrieved successfully.