How do I create a Studio Express-1 avatar? | Synthesia Knowledge Base

How do I create a Studio Express-1 avatar?

Learn how to film, submit, and create a Studio Express-1 avatar in Synthesia.

Written by Meg Farley
Updated today

What Is a Studio Express-1 Avatar?

Studio Express-1 avatars use Express-1 technology, which means the avatar automatically adjusts its facial expressions based on your script. This makes the output more dynamic and natural than a standard recorded avatar.

💡 Want to understand how Studio Express-1 avatars compare to other avatar types? Check out Understanding Synthesia's avatar types: stock, custom and customizable

Synthesia Requirements

The quality of your avatar depends on Synthesia receiving footage that meets these guidelines. If the standard is not met, a reshoot may be required.

💡 Warm up before your first take. Practice hand gestures, settle into a natural stance, and read your script out loud once — it makes a difference.

1. Video Background

The background of your footage must allow Synthesia to cut out the actor from the video.

Use a green screen background, well lit with no shadows. This is required for the highest quality avatar.
Make sure the background does not clash with the actor's clothing or skin tone.
Do not wear green clothing against a green background — use a white or blue background if the actor is wearing green. Footage may be rejected if this is not followed.
Do not remove the background yourself. Synthesia handles this before avatar creation.

✍️ Studio Express-1 avatars are not delivered with the original background.

During the avatar creation process, Synthesia removes the background behind the performer. Your avatar will appear without any background when used in a video project. This is expected and is part of how the avatar is built — it allows the avatar to be placed into any scene or template in Synthesia.

💬 What do I do if I can't shoot in a studio?

Footage shot in an office can be used, but ensure the actor stands out from the background and that you have good lighting and clear audio. Make sure you have access to reshoot footage based on Synthesia's feedback, if required.

2. Camera

A good quality camera is critical. Use a dedicated video camera — do not use webcams, streaming software, or applications that capture recording over an internet connection.

Camera requirements:

UHD resolution at 3840x2160
29.97 fps or 30 fps
Camera positioned at the same height as the actor's eyes
Sharp focus on the face
Standard video production technique using a camera and external microphone is required.

💬 What if I can't shoot at UHD?

HD footage is acceptable, but make sure to shoot the exact framing you would like to see for the avatar, as adjustments cannot be made during the creation process.

💬 What if I am shooting video in Europe?

If you shoot with conventional room lighting such as office fluorescents, shoot at 25 fps to avoid video flicker. If you have studio lighting, record at NTSC 29.97 fps.

3. Framing

Frame the footage to show the upper body of the actor.

UHD (3840x2160) is recommended — Synthesia can reposition the framing if needed.
If filming in HD (1920x1080) frame the avatar exactly as you want the final avatar to appear. Framing of HD footage cannot be adjusted without losing image quality. Ensure the actor's hands remain outside of the frame at all times.

Examples of correct upper-body framing

4. Lighting

Maintain a consistent lighting setup throughout filming. Synthesia does not offer post-production adjustments. Ensure proper lighting during recording.

Use a three-point lighting setup (a two-point setup is also acceptable).
Ensure fixed, even illumination on the actor.
Avoid shadows, particularly across the face.

Three-point lighting setup

Use a Key light at the front to illuminate the actor, positioned slightly to one side to add depth to the face. Use a Fill light on the opposite side to reduce shadows, and a Backlight to separate the actor from the background.

Two-point lighting setup

Use a softer Key light to create contrast from left to right. Note that this can limit the depth of the subject if the right equipment and intensity are not used.

💬 What if a two-point or three-point setup is not possible?

Ensure the actor is well lit from the front and clearly stands out from the background. Check that there are no strong shadows across the face or body. Keep the light flat and even.

5. Audio

✍️ Voice clones are now included with Studio Express-1 avatars. Ensure all audio is clean and free of off-camera noise.

Synthesia uses audio to train AI algorithms to match lip movement. On-camera audio is acceptable as long as it is clear.

Audio requirements — footage may be rejected if these are not met:

Audio must be in sync with the video.
No echo.
No background noise.
Only the actor should speak — do not feed lines to the actor during recording.
A lapel mic (lavalier) or boom mic is ideal.

✍️ Synthesia cannot remove a visible lapel mic. Hide the mic carefully and practice before filming to ensure it does not scratch against clothing during movement.

💬 Do I need perfect audio?

Yes. Avoid all background noise and ensure no one other than the actor is speaking, especially at the same time. Synthesia requires 3 clear takes of the actor speaking to camera with as little background noise as possible.

6. Color

🚨 Synthesia does not adjust the color grade for the actor during the avatar creation process. Provide footage with the look you want your avatar to have.

Avoid dark footage — ensure good lighting throughout.
Shoot with the desired look on camera, or grade the footage before submitting.
Do not provide raw, ungraded footage.

7. Wardrobe

Dress the actor in the clothing you want the avatar to wear.

Do not wear clothing that clashes with the background.
Avoid green clothing.
Makeup must look natural.
Keep hair tight to the head — avoid gaps where the background is visible through the hair.
No sunglasses — the eyes must be visible.
No low hats — the forehead above the eyebrows must be visible.
No long beards — short beards are acceptable, but check with customer support before filming to confirm.
Tattoos are generally acceptable — check with customer support before filming to confirm.

💡 To achieve the best footage, avoid the following:

Strong shadows.
Camera blur.
Not speaking to camera.
Background action.
Hair across the face.
Sunglasses.

If in doubt, ensure the actor's face is clearly visible to the camera at all times.

8. Glasses

Studio Express-1 technology works with glasses, though some frame types can be more challenging for the AI model. To minimize risk, record takes both with and without glasses.

When recording with glasses:

Ensure the frames are not positioned close to the performer's eyelids.
Avoid excessive head rotation or tilting.
Avoid see-through frames.

✍️ Submit glasses and non-glasses footage separately. Create two separate submissions on the platform — one set with glasses and one without.

9a. Performance Demonstrations

The performance of the actor on camera defines how your avatar videos will look on Synthesia.

Express-1 Avatars are expressive but also emotive. Watch our demo videos below for more guidance on performance.

Performance Requirements

Performance guidance

9b. Performance - (Head & Eyes)

The below instructions should be followed for optimal results:

The Performer should keep their eyeline towards the camera.
Head should move NATURALLY and be EXPRESSIVE.
No sneezing or coughing. Take must be redone if so.
If reading from a teleprompter, make sure not to have it angled far from the camera.
The performer should not be looking too far away from the camera.
Do not adjust the distance between the performer and the camera throughout.
Performer should not rock forwards and backwards or move closer to the camera.

9c. Performance - (Hands & Arms)

For an avatar with great movement in the upper body and arms, we would recommend the following:

Hands should be kept at waist level at all times. Your avatar may not be created if this is not followed.
Move your shoulders and head naturally as you read the script.

9d. Performance (Emoting)

The performer does not need to be a professional actor.

We require pronounced and consistent facial expressions:

When the direction is to be happy, smile, maintain professionalism and calmness, and ensure to smile.
When the direction is to be sad, portray negativity, anger, and upset emotions.
When the direction is to be excited, exhibit excitement, happiness, and positivity. Enjoy the moment!

💡 DIRECTING TIPS 💡

For EXCITED, ask your performer to:

Persuade.
Convince.
Seize The Moment.
Share a discovery.
Encourage.
Motivate.

For SAD, ask your performer to:

Demand.
Challenge.
Moan.
Complain.
Confront.
Warn.

10. Script

Each video take must contain a reading of the full performance script. This should be repeated 3 times for 3 separate takes. The script will direct the performer. Make sure to follow emotional directions throughout. Download the Performance script for Studio Express-1 Avatar here.

✍️ Each performance should follow the same emotional direction. The avatar will mimic the performance - if the performance is not expressive or emotional enough, this will be reflected in the avatar.

💬 What if the actor is not a native English speaker?

The script can be translated into other languages, but please do keep the content the same. It is important to deliver a performance that has sections for happy, sad and excited emotions.

💬 Should I cut the video clips to provide the exact script?

No, please do not cut the videos. We require full takes without any cuts. It does not matter if the actor gets the lines wrong or word skipped. Just ask the actor to continue delivering the script naturally even if they make a mistake.

💬 Do I need to use a teleprompter?

You will get the best result if the actor is familiar with the script and then delivers to camera with a teleprompter to assist. If you don’t have a teleprompter then use a tablet positioned as close to the camera as possible, but please do check the eye line.

11. Consent Recording

As part of the submission we require a recording of the performer reading the following script to camera. This script needs to be read in the actor’s native language, for which we provide translations. For this recording you don’t need to worry about the style of the delivery, as long as the script is clearly stated by the performer, while facing the camera.

Download the consent script: Synthesia Consent Recording script.

12. Submitting Your Footage

Once you have all recordings, submit your footage via the Studio avatar portal in Synthesia.

Go to the Avatar section in Synthesia.
Select the Studio avatar option.
Follow the on-screen instructions to upload your files.

Submit the following:

3 performance video takes (each containing a full reading of the performance script)
1 consent recording

File requirements:

Video length is a minimum of 2 minutes
.h264 codec
.mp4 format
Less than 2 GB per video
29.97 or 30 frames per second
UHD 3840x2160 resolution

✍️ Footage must be a continuous take with no jump cuts or mid-take edits. If any part of the performance is removed, Synthesia will not be able to use it to create your avatar.

By submitting your footage, you confirm that the actor understands their likeness will be used to create an AI avatar and that you agree to the Synthesia Ethical Guidelines and Synthesia Terms and Conditions of Service.