Overview
Character Animator takes a layered artwork file and turns it into a puppet that performs from a camera and a microphone. Head position, eye direction, eyebrow movement and mouth shapes are driven by a face in front of the webcam, and the lip sync is generated from the audio in real time rather than being keyframed phoneme by phoneme.
The puppet rigging is done through layer naming rather than a bone editor, which is a deliberate and effective simplification. Name a layer for the left eyebrow and it becomes the left eyebrow. Behaviours are then attached on top: dangle for hair and loose clothing, physics for props, walk cycles for legs, breathing for idle motion, and triggers that swap a limb or an expression on a key press.
The live output is what separates it from a general animation package. The performance can go straight out to a streaming or broadcast application while it happens, which is why it turns up on live shows and on stream overlays rather than only in finished cartoons. For recorded work, takes are captured and layered, so the mouth can be reperformed over an existing body take, and the result sends onward for finishing with keyframes and effects.
What it does well
Performance capture
A webcam drives head, eyes, eyebrows and mouth shapes, with lip sync generated from the microphone audio as you speak.
Rigging by layer names
Puppets are built from a layered artwork file with named layers, so a rig takes minutes rather than an afternoon in a bone editor.
Behaviour library
Dangle, physics, walk cycles, breathing, blinking and idle motion applied as behaviours rather than animated by hand.
Triggers and swap sets
Bind expressions, hand positions and props to keys, with swap sets that cycle through variants during a take.
Live output
Send the performance out to a broadcast or streaming application in real time, which makes it usable on a live show.
Changes in this build
- Face tracking improved in low light and with partially obscured faces.
- Lip sync generation reworked with better handling of fast speech.
- New behaviours for secondary motion on props and clothing.
- Take recording made lighter so long sessions do not fill the project.
- Body tracking accuracy improved for arm and torso movement.
What is in the package
- Character Animator 2026 offline installer, 64-bit
- Starter puppet library with editable artwork
- Behaviour and trigger preset library
- Live output components for broadcast applications
- Multilingual interface files, 12 languages
System requirements
| Processor | Intel or AMD 64-bit, six cores recommended |
| Memory | 16 GB minimum, 32 GB for complex puppets |
| Graphics | GPU with 4 GB VRAM |
| Storage | 12 GB free plus space for takes |
| Display | 1920 x 1080 minimum |
| Capture | Webcam and microphone for performance capture |
Installation
- Extract the archive to a local drive.
- Close any other Adobe application before running setup.
- Run the installer and choose the interface language.
- Connect the camera and microphone before first launch so they are detected.
- Open a sample puppet and calibrate the face tracking once.
Before you start
Even, front on lighting improves face tracking far more than an expensive camera does.
Puppet artwork has to follow the layer naming convention or the rig will not bind.
Turn off automatic updates so the installed build stays where it is.
Questions about this title
Do I need a special camera?
No, an ordinary webcam is enough. Lighting matters more than the sensor.
Is lip sync automatic?
Yes, mouth shapes are generated from the audio and can be corrected afterwards.
Can I stream with it?
Yes, the live output sends the performance to a broadcast application in real time.
Where do puppets come from?
From a layered artwork file with named layers, or from the included starter library.
About this listing
This entry was checked on a clean install of Windows 10 / 11 (64-bit) before it was published, and it is rechecked whenever the package is rebuilt. The figures on this page come from the site index rather than from the publisher, so the download count is what people here have actually pulled.
If a mirror stops answering, use the contact form and it gets replaced. Reports about a specific title are handled faster than general messages because they name the file, the mirror and the point at which the transfer stopped.
Requests for a different version, a different language build or an older release go on the requests page. Older versions of a title are usually still held even when only the current one is listed.