Record or upload
Use a short sample with one speaker, minimal echo, and no music.
- Clear speech
- Up to 10MB
Clone a voice online from a short audio sample, then generate natural text-to-speech voiceovers with flexible AI controls.
Generated audio will appear here.
A clean reference recording gives an AI voice clone its character. Record naturally in a quiet room, then use the same cloned voice whenever you need consistent text-to-speech narration.

Use the workspace above to move from source audio to a usable voice.
Use a short sample with one speaker, minimal echo, and no music.
Only create a clone of your own voice or a voice you are authorized to use.
Add text, choose your cloned voice, then listen before downloading your audio.
Avoid conversations, overlapping voices, music, and sound effects.
Speak naturally instead of rushing or exaggerating each word.
A quiet space helps preserve tone and pronunciation in the cloned voice.
Generate a few sentences first, then adjust your source sample if needed.
Use the same voice identity for narration in supported languages.
Guest reference audio, voice clones, and generated audio are automatically time-limited.
Guest use is available with configured usage limits. Your guest voice assets expire automatically after the configured period.
A short sample of clear, continuous speech works best. Aim for three to fifteen seconds when recording in the browser.
Only if you have clear authorization from the voice owner. You must confirm this before creating a clone.
The guest clone, reference audio, and its generated audio are no longer available and are cleaned up automatically.