Get Supported Voice For Audio Models

Fetch the supported voice_id values and the default voice for TTS models before calling audio generation. Where an entry has voice_description: true (Qwen/Qwen3-TTS-VoiceDesign), the model takes a description of the voice as voice_id instead; where it has voice_description: false and lists no voices (ResembleAI/chatterbox-turbo), the model has its own voice and takes no voice_id. Where given, max_characters is the most characters one request takes.

GET https://api.oneinfer.ai/v1/ula/get-supported-voice-for-audio-models?model_names=bulbul:v3
Authorization: Bearer <YOUR_TOKEN>

{
  "api_details": {
    "api_status": "success",
    "api_message": "API has returned response successfully."
  },
  "data": {
    "by_model": {
      "bulbul:v3": {
        "provider": "sarvam",
        "default_voice_id": "shubh",
        "voice_ids": ["abhilash", "anushka", "manisha", "vidya", "shubh"]
      }
    },
    "unknown_models": []
  },
  "error": {}
}