In the HeyGen AI Studio, you can customize your script's audio by adjusting the voice to match your individual projects. Whether you're crafting a professional explainer, a lighthearted promo, or an internal training video, selecting the right voice helps set the tone for your message. You can choose from a wide range of voice options across different accents, genders, and styles — all designed to bring your script to life. Each voice is AI-generated and optimized for clarity, making it easy to find one that aligns with your content’s purpose and audience.
🛎️ Note: New to Voices? This article would be a good place to visit after you've already checked out our How to Get Started with Voices article.
How to Use a Voice in AI Studio
Step 1: Start Your Project
First, open AI Studio to start your project.
Step 2 : Select a Voice or Change the Assigned Voice
You can choose, create, or connect a voice directly in AI Studio.
Go to the Voice section in the right-side menu. From there, you can:
Pick a voice from our Voice Library
Create a new voice
Design a voice with AI
Use an integration to connect a third-party voice
Once selected, apply the voice to your avatar and preview the result before generating.
Apply Voice to Other Scripts: To keep things consistent for your whole project, use the 'Apply to all scenes' button, which assigns your selected voice to every script in your project.
Adjusting Voice Settings
You can fine-tune your voice settings to better match your project’s needs.
In AI Studio, you can adjust settings such as volume and speed to get the right sound for your video.
Please note that pitch adjustment such as similarity, stability and style is only available for specific engines (ElevenLabs + ElevenLabs Multilingual v2 or Fish).
Advanced: Voice Engine Selection
You can also choose the voice engine that best fits your use case:
Auto — HeyGen automatically selects the best engine for your video.
ElevenLabs — Natural, studio-quality narration in many languages. Available in V2 and V3.
Panda — HeyGen’s expressive engine, supporting Voice Director and Voice Mirroring for more emotional nuance and precise timing.
Starfish — Optimized for Chinese, Japanese, and Korean.
Fish — Allows you to select different accents and locales for more region-specific English pronunciation and delivery. Powered by Fish.audio for highly expressive speech generation.
Want more control over your voice? check out Voice Director and Voice Mirroring here!
Improve Your Voice
In AI Studio, you can refine your AI voices by selecting their current voice, prompting specific modifications in a chat, and choosing between an automatic fix or a manual re-recording.
Once the AI generates the enhanced version, it can be saved to permanently replace the original voice, ensuring a more professional and personalized vocal output for all future projects.
Quick Wins: Improve Voice Quality with Simple Tweaks
If you're experiencing pronunciation, pacing, or consistency issues, here are a few quick tips that often improve the final result:
Most voice problems have a fast fix. Match your symptom to the first step below — they're ordered easiest-first, so try them in order and stop when it sounds right.
Sounds flat or monotonous
Switch to a more expressive engine (ElevenLabs V3 or Panda), or run Voice Remix. Still flat? On ElevenLabs V2, lower Stability and raise Style Exaggeration — lower stability means more emotion.
Cloned voice doesn't sound like me
Re-clone with background noise removed (it's removed by default, so skip if already done). Then nudge the Similarity slider up in steps — 75 → 80 → 85 → 90 → 95 — until it locks in. For English, the Fish engine often clones closest.
Audio quality is rough
Re-clone with a cleaner recording, then run Voice Remix → "Improve Audio Quality" preset.
Accent is wrong
For English with a specific accent, use Fish; for other languages, use ElevenLabs V3. You can also Remix to add the accent, or set the accent explicitly in settings.
✨ The single biggest fix: re-clone with more, cleaner audio. Almost every voice issue improves when the engine has better source material to learn from.
Preview Your Voice
Before rendering your scene, use Audio Preview to review your narration, pronunciation, pacing, and timing. You can preview your voice by clicking the Play button directly on the desired scene or from the timeline.
🛎️ Note : For longer videos, split your project into shorter scenes (around 1–2 minutes) to make editing and previewing more efficient.
⚠️ Audio Preview is free but excessive audio previewing over extended periods may temporarily trigger usage safeguards.
A new Audio Preview is generated when you make changes that affect the narration, such as:
Editing the script, even a single character
Switching to a different voice
Adjusting voice settings
Changing anything else that affects the narration
After any of these changes, clicking Play generates a new Audio Preview for the entire scene.
Tip: You can also manually regenerate the audio at any time by clicking the three-dot (⋮) menu on the scene and selecting Regenerate Audio. This is useful if you'd like to refresh the narration after making changes.
Previewing the voice will not generate lip sync, facial expressions, or avatar animation. To see how the avatar moves and speaks in the final video, you’ll need to render the scene, which uses credits. You can learn more in our Preview & Render article.
Frequently Asked Questions (FAQ):
What do the different Voice Models Mean?
When using your custom voice on HeyGen, you can choose which ElevenLabs voice model powers it. By default, we use Multilingual v2, a solid all-purpose model that works well across different languages. If you’re looking for faster speaking speed, consider switching to one of the Turbo models—just note that some of them are English-only.
For more information please view this article from ElevenLabs on their different models.
Can I translate my script to be read in another language using one of the Voices?
Yes! You can write your script in any language and seamlessly translate it into another language directly within the Text Editor.
You also have the option to translate an existing video after it has been created. Simply open the video from the Share Page and select the translation option to create a new draft in your desired language.
Can HeyGen or third-party Voices be used in translation?
AI-generated voices are available for use in the translation feature for Team and Enterprise Users via Proofread.
How can I integrate my ElevenLabs [or other third-party] Voice into HeyGen?
HeyGen supports the integration of your ElevenLabs, LMNT, Play, or Character AI Voice, allowing you to enhance your videos further with advanced third-party voice options. To learn how to integrate your third-party voice, check out this article.