Generate Voice Candidates

This endpoint generates three voice candidates based on your text description.

Request

POST https://app.resemble.ai/api/v2/voice-design

Request Body

Parameter	Type	Required	Description
user_prompt	text	Yes	Text description of the desired voice characteristics

Example Request

curl --location 'https://app.resemble.ai/api/v2/voice-design' \
--header 'Authorization: Bearer YOUR_API_TOKEN' \
--form 'user_prompt="A middle-aged female with an Australian accent, friendly and approachable, like a knowledgeable tour guide"'

Response

Response Body

Field	Type	Description
voice_candidates	array	Array of voice candidate objects
voice_candidates[].audio_url	string	URL to the audio sample
voice_candidates[].voice_sample_index	integer	Index of the voice sample (0, 1, or 2)
voice_candidates[].uuid	string	UUID of the voice design model

Example Response

{
    "voice_candidates": [
        {
            "audio_url": "<audio_url>",
            "voice_sample_index": 0,
            "uuid": "<voice_design_model_uuid>"
        },
        {
            "audio_url": "<audio_url>",
            "voice_sample_index": 1,
            "uuid": "<voice_design_model_uuid>"
        },
        {
            "audio_url": "<audio_url>",
            "voice_sample_index": 2,
            "uuid": "<voice_design_model_uuid>"
        }
    ]
}

Notes

• The system will always return 3 different voice samples.
• The uuid is required for the next step - Create Voice from Candidate.
• The uuid will be the same for all 3 samples since they belong to the same voice design request. The voice_sample_index is what distinguishes between the different samples.