HeyGen Voice is our new voice model, and the first built to adapt to video. It ranks first on Artificial Analysis’ voice leaderboard. It delivers your script in your voice, with your pacing and expression, while adapting to the scene’s acoustics. It’s included in every HeyGen video at no extra cost.

What you can do

Create your voice from a short recording

Upload up to 60 seconds of audio from any device, including your phone. HeyGen Voice learns your timbre, accent, and phrasing, keeping your voice and dropping the phone quality.

Design a voice from a picture

No recording? HeyGen Voice designs a voice for the person and place in an image.

Match your voice to the scene with Adaptive Mode

HeyGen Voice reads your scene, including the room, the mic distance, and the delivery the shot calls for, and fits your voice to it.

Keep one consistent sound with Exact Mode

Exact Mode keeps your voice and your recording conditions together, so every new line sounds like it came from the same session. A ten-minute script sounds like one take, not forty stitched together.

Add energy with Expressiveness Boost

One slider adds emphasis, rhythm, and inflection while still sounding like you. Set it to zero to return to your baseline delivery.

Fine-tune the scene

Pitch variation, background level, and ambient steadiness each have their own slider so the same scene can sound nearly clean or clearly busy.

Use your voice in other languages

The same recording can read Spanish, German, French, or Chinese with a native accent.

Go further with Professional Voice Clone

Train HeyGen Voice on 30 minutes to two hours of your recordings so it learns finer details, like how your accent moves across words. Available as an add-on.

Build with the API

Generate speech from your professional voice clone with one REST request, or stream audio as it’s generated, with optional word timestamps for captions. It’s $30 per million characters, with no packages or minimums.