Skip to main content

How to Use Voice Mirroring and Voice Director

Try these new voice tools to fine-tune your avatar’s speech, allowing for more natural and emotionally aligned video narration.

Emily avatar
Written by Emily
Updated over 2 months ago

Voice Mirroring and Voice Director are powerful tools that give you full creative control over your AI avatar’s tone, pacing, and emotional delivery. Whether you want your avatar to sound excited, calm, or even sarcastic, these features let you move beyond basic punctuation and pauses.

With Voice Mirroring, you can record your own audio and have your avatar mimic your vocal style, unique accent and intonations, but in their voice.

With Voice Director, you can guide your avatar’s emotional delivery line by line, like directing a voice actor in real time. In this guide, you’ll learn how to use both tools to create more natural, expressive, and on-brand AI videos.


Voice Mirroring vs. Voice Director

Sometimes, text-to-speech (TTS) alone doesn't capture the subtle tone, pacing, or emotion you envision. That’s where these two powerful tools come in:

Feature

Best For

How It Works

Level of Control

Requires Recording?

Voice Mirroring

Personal tone, emotional depth, custom delivery

Record/upload your voice and have it mirrored by an avatar

High – full nuance and emotion

✅ Yes

Voice Director

Fast, efficient control over tone/emotion

Add tone/emotion instructions in text

Medium – flexible and quick

❌ No

Use Voice Mirroring when you want a precise, realistic, and humanlike delivery that matches your own vocal style.

Use Voice Director when you want to shape the emotion and tone of a script without recording anything manually.


How to Use Voice Mirroring

Interested in a Video Tutorial? Check out our Perfecting your AI Video Scripts & Pronunciation (Part 2) video below. Skip to 1:57 to learn about Voice Mirroring.

Use in a New Scene

Create a new video or open an existing project, add a new scene and choose an avatar, then in the scene editor click Audio, where you can either record new audio or upload an existing file — for this example, we’ll choose ‘record audio’.

Record the audio using the desired tone, emotion and pacing. Then Upload, enable Voice Mirroring ✔, and click Add to Scene. HeyGen will automatically transcribe your audio, display the transcription in the scene, and allow you to review it for accuracy.

Select a public avatar and voice from HeyGen’s library, and your recorded delivery—tone, pacing, and emotion—will be mirrored onto the selected voice. For high-quality audio narration in HeyGen, check out our full guide on using HeyGen AI Voice.

Use an Existing Scene

If you initially create a script with text only and find that the verbal delivery doesn’t sound the way you want, you can refine your results with Voice Mirroring.

Open the scene with the existing text script and click Convert to Voice Mirroring. HeyGen will automatically display your script like a teleprompter, and you can record or upload your voice with your desired pacing, tone, and emotion. Your avatar will mirror your performance, resulting in a more natural and authentic delivery.

This workflow lets you quickly improve the expressiveness of a script without needing to rewrite or manually adjust voice settings.

Tips for Best Results

  • Use natural, clear instructions in the tone field.

  • Try different presets to see how the voice and facial expressions adjust.

  • Write your own direction for unique or brand-specific delivery.

  • Great for fast iterations without needing a microphone.


How to use Voice Director

Interested in a Video Tutorial? Check out our Perfecting your AI Video Scripts & Pronunciation (Part 2) video below. Skip to 2:40 to learn about Voice Director.

In the Editing Studio, look for the megaphone icon 📢 on the left-hand toolbar. Clicking it will open the Voice Director panel.

First, if you are not using the Panda Voice Engine, you will be prompted to switch. You can always switch back in your settings later.

Next, you’ll see a list of preset tone options: Excited, Casual, Calm, Cool, Serious, Funny, Angry, Sarcastic.

Select the one that fits your message best.

Each tone comes with a default text instruction, such as "speak in a calm, steady tone with a warm vibe." You can edit this prompt to match your unique delivery needs.

💡 Pro Tip: Want something more specific? Try writing your own:

  • “Slow and thoughtful, like explaining something serious.”

  • “Quick, upbeat, and energetic—like a YouTube intro.”

  • “Deadpan and dry, almost sarcastic.”

Once you're happy with the direction, click Save. Then hit Play in the preview window to hear your avatar perform the line with your chosen delivery.

You can always go back and adjust until it feels right.

Supported Voices for Voice Director

Voice Director works with most voices in HeyGen, including:

  • All custom voice clones created on HeyGen

  • Public voices from ElevenLabs and OpenAI

  • Imported voices via ElevenLabs API (including pro voices)

Tips for Best Results

  • Emphasize the tone and rhythm you want reflected.

  • Avoid overacting unless your script calls for it.

  • After recording, you can convert the audio into a script, so you can re-record with a teleprompter for even better control. Check out how to write and use engaging video scripts in HeyGen.

  • Start with short segments to get familiar with pacing and processing speed.


With HeyGen’s Voice Mirroring and Voice Director, you can now direct exactly how your avatar delivers your message—adding human-level nuance that text input alone can't achieve. Record, mirror, and deploy more authentic, engaging AI videos faster than ever.

Ready to get started? Head to your HeyGen dashboard and experience the future of avatar communication with Voice Mirroring and Voice Director!

Did this answer your question?