|You can find and manage voices in three places: AI Studio when working inside a project, the Avatars page when assigning or setting a primary voice, and Proofread. You can create a new voice from any of these locations.
For this walkthrough, we’ll start in AI Studio.
Start creating a new voice in AI Studio
Open an existing project or create a new one in AI Studio. In the script panel, click on the current voice to open the voice menu. From there, select Add New Voice.
You’ll see two options: integrate a third-party voice provider or create a new voice directly in HeyGen. For this tutorial, choose Create New Voice.
Upload a voice recording
You’ll be prompted to upload a voice recording. This recording is used to build your voice clone, so quality is important.
Use a high-quality external microphone, a Bluetooth mic, or a modern smartphone. Avoid using your laptop’s built-in microphone. If you’re using an external mic, keep it about 6 to 8 inches from your mouth, make sure it’s unobstructed, and avoid contact with clothing or fabric.
Record in a quiet, noise-free environment. Background noise can reduce clarity and consistency. Speak naturally and clearly, with some emotional variation. Talking about familiar topics such as your daily routine, hobbies, or personal stories helps capture your natural tone and energy.
If you’re reading from a script, avoid sounding flat. Use pauses, vary your pacing, and maintain a steady, conversational volume.
Generate and refine your voice
Once your recording is uploaded, HeyGen will generate your voice clone. If the result isn’t quite right, you can refine it using Voice Doctor.
Click Enhance Voice to adjust settings such as accent or voice engine, or describe what feels off, for example, if the voice sounds too flat or doesn’t quite match your natural tone. Voice Doctor will generate improved options that you can preview.
When you find a version you like, you can save it as a new voice or replace the existing one. If none of the options feel right, you can continue refining or try a different approach.
Create a voice with Voice Design
If you prefer not to record yourself, you can use Voice Design instead. Click Voice Design, enter a prompt describing the voice you want, and HeyGen will generate several options.
You can select one, regenerate additional options, or fine-tune the prompt until the voice fits your needs.