Voice Guide

Hume’s text-to-speech (TTS) API lets you specify which voice to use when synthesizing speech. You can use a custom voice that you have saved or select one from Hume’s Voice Library. If you omit the voice field, the model will generate one dynamically based on the input text and optional description.

This guide explains how to specify a voice in both standard and streaming TTS requests.

Voice reference options

You can specify a voice in your request using either its id or name. Each voice belongs to a provider, which indicates the source of the voice and who can access it.

The provider field accepts the following values:

If you omit the provider field, it defaults to CUSTOM_VOICE. To use a voice from the Voice Library, you must explicitly set the provider to HUME_AI.

1
{
2
  "voice": {
3
    "name": "My Custom Voice",
4
    // "provider": "CUSTOM_VOICE" (optional)
5
  }
6
}

1
{
2
  "voice": {
3
    "id": "795c949a-1510-4a80-9646-7d0863b023ab",
4
    // "provider": "CUSTOM_VOICE" (optional)
5
  }
6
}

Specify a voice in your request

To specify a voice for speech synthesis, include the voice field in the first utterance of your request. That voice will be used for all subsequent utterances unless you override it in a later utterance.

Both standard and streaming TTS endpoints support voice selection. The request body format is identical across both.

1
curl https://api.hume.ai/v0/tts \
2
  -H "X-Hume-Api-Key: $HUME_API_KEY" \
3
  --json '{
4
  "utterances": [
5
    {
6
      "text": "Beauty is no quality in things themselves: It exists merely in the mind which contemplates them.",
7
      "voice": {
8
        "id": "9e068547-5ba4-4c8e-8e03-69282a008f04",
9
        "provider": "HUME_AI"
10
      }
11
    }
12
  ]
13
}'

Resources