OpenVoice

About OpenVoice

OpenVoice is a versatile instant voice cloning model that allows fine-grained control over speaking style. Unlike other cloning models, OpenVoice separates voice identity from speaking style, allowing you to take a cloned voice and apply different tones - cheerful, sad, angry, excited, or whispering - without new reference audio.

Key Features

Use Cases

How to Use OpenVoice

1. 1

   Sign up free or try the demo

   Create a free TextToSpeechAI account to get starter credits, or use the on-page demo to hear OpenVoice before committing. No local GPU or install is needed - everything runs on our servers.

2. 2

   Choose OpenVoice and upload a reference clip

   Select the OpenVoice engine, then upload a few seconds of clean reference audio to instantly clone the target voice. OpenVoice captures the speaker identity so you can reuse it across any text and tone.

3. 3

   Enter your text

   Type or paste the script you want spoken in the cloned voice. OpenVoice supports around 10 languages and cross-lingual delivery, so you can write in a different language than the reference clip.

4. 4

   Pick a tone style and generate

   Choose one of the nine OpenVoice tone styles - default, friendly, cheerful, excited, sad, angry, terrified, shouting, or whispering - then generate. The same cloned voice will speak with that emotional delivery.

5. 5

   Download or use the API

   Download your audio as MP3, WAV, or OGG, or automate generation through the TextToSpeechAI API by passing your cloned voice and tone style in each request.

OpenVoice API

Generate speech programmatically using the TextToSpeechAI REST API.

curl -X POST "https://api.texttospeechai.com/v1/generate/" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "OpenVoice can speak in any tone \u002D cheerful, sad, or even whispering.",
    "voice": "en_US-lessac-medium"
  }'

Frequently Asked Questions