All Collections
How to create and edit an Avatar
HeyGen Studio Avatar - Instruction
HeyGen Studio Avatar - Instruction

Please follow the following instructions when sharing footage to ensure the perfect results for your HeyGen Studio Avatar.

Yuumi avatar
Written by Yuumi
Updated over a week ago

Thanks for your interest in HeyGen Studio Avatar. Studio Avatar is exclusively available for enterprise customers to customize.

The quality of your Avatar will highly depend on the footage you provide.

Technical Setup

Lighting

  • Make sure there is plenty of light

  • Avoid strong shadows on the face and body

  • It is worth using professional operators and lighting services. We highly recommend following 3-point lighting(https://www.youtube.com/watch?v=IXe3rEWrDVg)

Backgrounds

  • Footage must be recorded in front of a green screen. We will then have our team do studio-level video matting for you

Camera

  • The camera should film in Ultra HD (4K) and support 60FPS

  • Set the framing from the waist or the mid torso up

Audio

  • Use a special lapel or Boom mic that clearly captures your voice

  • Make sure there is no background noise or interruptions

Footage Delivery

  • Provide each take as a separate file

  • Do not edit and cut the footage

You can do the recording yourself, but we recommend you do the filming at a professional studio. You can share the setup instructions with them.

Preparation

Script

Please prepare at least two scripts yourself. You can talk about anything you are interested in or familiar with. Check the following topics for reference:

  • News

  • Self-introduction

  • Travel experience

  • A story

  • Your plan for the weekend

  • Your favorite food / movie / book

Look and Performance

Here is an example of good footage and speech performance:

Look

  • Clothes, haircuts, and makeup will stay the same as in your footage. We won’t be able to change the look later. So please make a recording in suitable clothing that will fit most of your videos, or choose a casual look that is suitable for different types of videos.

  • Please avoid reflecting outfits that might directly emit or reflect green screen background.

  • Avoid loose hair/ reflective accessories / long beards / hat / sunglasses.

Performance

  • Keep your eyes on the camera and avoid large head pose for the whole duration of the speech.

  • Be natural and calm. Your avatar will perform exactly the same way as you do.

  • You can use any text you want in any language, the more important is how you perform.

  • Perform the same as you do in public or filming a regular video speech: your movements should be natural, and feel free to use a few gestures and facial movements.

  • We require at least two 2-minute long (4 minutes in total) uninterrupted speeches.

  • Pausing with lips closed is one of the most important parts of the performance. Make sure you have.

    • 2-4 second pause (Lips closed) at the beginning

    • 2-4 second pause at the end of each paragraph

    • 1-3 second pause in the middle between some sentences

The Consent Statement

  • HeyGen Avatar is a photo-realistic representation of a human, which can be built with the person's explicit consent only.

  • A formal consent statement must be recorded and also sent along with the above footage. An example is given as follows:

    • I, John Doe, allow HeyGen to use the footage of me to build a HeyGen Avatar for use on the HeyGen platform.

Submitting your footage

  • Please upload your original footage to a cloud storage service like Google Drive, and share the link with ­[email protected]. If you have any questions, please contact us anytime at: [email protected]

Frequently Asked Questions

Can the presenter use hand gestures?

  • Yes, natural and general gestures are acceptable as long as your gesture does not overlap with your face, please keep your face always fully visible to the camera without too much head movement.

  • However, we suggest avoiding exaggerated hand gestures and directional gestures, such as pointing or counting, during the whole performance , since these may be hard to match the speech in the video creation.

Did this answer your question?