Adobe Character Animator can send a live puppet scene to a receiving broadcast application or video device through its Stream workspace and Live Output preferences. The important detail is that the receiver gets the current Scene panel output, not an automatic copy of your whole project or its background-colour setting.
A reliable setup is a short chain: prepare the puppet and scene, choose an output path, start streaming, then check what the receiving application actually sees and hears. Rehearse the full route before using it on YouTube, because framing, alpha handling and audio sync depend on the receiving setup as well as Character Animator.
Prepare the puppet and scene
Open or create the puppet you intend to perform with, then place it in a scene. In Character Animator, a scene can contain puppets, audio files and recorded takes; it is the working composition that you will send live. The Adobe overview of scenes is useful if you are still deciding how to organise a project.
Before selecting an output, check that the intended puppet is in the scene and that its visible position works within the Scene panel. Think about the composition the audience needs: a full-body character may need more space around it than a talking-head puppet, and a lower-third or title should be placed inside the panel rather than assumed to appear from somewhere else in the project.
Character Animator uses inputs such as a camera, microphone and mouse to control a puppet. Calibrate the camera tracking rest pose before a performance so the puppet starts from a sensible neutral position. Check the microphone input and speak at the volume you expect to use on air. Adobe's workspace guide describes the main workspaces and their roles.
For a single-character broadcast, Adobe says most USB microphones and audio interfaces can work. Choose a microphone that suits the room and rejects background noise, and use headphones when monitoring helps prevent speaker audio from being picked up again. A quiet room and a sensible microphone position can be more useful than changing settings later to compensate for fan noise, room echo or a distant voice.
Keep the performance plan simple at first. If you need a second puppet, prerecorded audio or additional scene elements, test those in the same scene you will send to the receiver. Do not rely on project-level assets or other open panels to act as part of the live picture; what matters is the current scene output.
Open the Stream workspace
Switch to the Stream workspace when the puppet and scene are ready. Its controls expose the live-output route and the Stream Live action at the bottom of the Scene panel. If you do not see the controls you expect, confirm that you are in the Stream workspace and that a scene is active rather than only viewing or editing a puppet.
Open Live Output preferences and choose the device or application path that matches the receiving setup. Adobe documents Mercury Transmit, NewTek NDI output and third-party video hardware as routes, depending on the installed setup. The exact device names visible on your system depend on its configuration, so do not assume a menu entry will be present merely because a tutorial shows it.
The receiving end must also be prepared. In a broadcast application, add or select the source that corresponds to the output path you chose, then position and size it in the broadcast composition. If your output goes to video hardware, make sure the destination is set to that input. Character Animator sending a signal and the receiver selecting it are separate tasks.
A useful rehearsal is to open the receiver first, add the intended source, and then start Character Animator's output. This makes it easier to distinguish an output-selection problem from a scene-framing problem. For broader broadcast planning, see this guide to choosing streaming software by workflow; the receiving application's role is to take the live source and compose it with the rest of your programme.
Choose Live Output preferences
Treat Live Output preferences as the hand-off point between Character Animator and the outside application or device. Select the relevant route, check any available output options, and then return to the Scene panel. The choice should reflect the gear and software you actually have installed, rather than a desired workflow that has not been tested.
Mercury Transmit is a documented output path for compatible video applications and devices. NDI is another option where the required NDI setup is available. Third-party video hardware can be appropriate when your broadcast chain already uses a capture or production device. Those choices are not interchangeable in every system: a receiver needs to be able to accept the format or connection you are sending.
If the output preference list does not show the destination you expect, check the connection and installed software or device support, then consult the current Adobe and device documentation. The research available for this article does not establish compatibility for every operating system, Character Animator version, NDI combination or capture device. Verify those details for your own setup instead of treating a sample configuration as universal.
When routing a YouTube programme through another application, keep the Character Animator output as one source among the receiver's other inputs. Your broadcast application may also handle titles, background plates, scene transitions and the final YouTube connection. If you need to plan access credentials separately, follow a careful process for transferring a YouTube stream key; do not place a stream key into unrelated source settings.
Select Mercury Transmit, NDI or video hardware
The best route is the one your receiver can reliably accept and that fits the number of machines in the performance. Mercury Transmit can suit a direct connection to an application or device that supports it. NDI can suit workflows that pass video between computers over a network. External video hardware can suit a production chain already built around physical video inputs. Confirm the receiver's documented support before committing to one.
| Output route | When it may fit | What to verify |
|---|---|---|
| Mercury Transmit | A receiver or device in the existing setup supports the output | The intended application or device appears and receives the signal |
| NDI | A one-computer or multi-computer workflow uses NDI-compatible endpoints | Both ends are configured for NDI, and the network carries the connection reliably |
| Third-party video hardware | The production workflow already accepts a hardware video input | The device is selected at the source and the matching input is selected at the receiver |
For a single-character performance, keep audio routing in view while choosing the video path. A microphone can feed Character Animator, the broadcast application, or a broader audio chain, depending on the setup. Avoid assuming that choosing a video output also routes the voice exactly as you want; confirm which application receives the microphone and which one sends the final audio to the stream.
For a multi-character NDI setup across computers, the network and audio plan matter as much as the picture. Adobe recommends wired microphones and a professional audio interface for each machine, along with Gigabit Ethernet over Cat 5e or Cat 6 between the machines and a switch. These are Adobe's workflow recommendations, not a guarantee that every network or interface will behave the same way. Test the actual devices and route in advance.
If the performance is on one computer, NDI may add complexity that you do not need. If separate machines make the performance easier to manage, network transport may be useful, but it adds another connection to diagnose. Write down which machine produces the puppet, where the voice enters, and which device sends the combined programme onward. That small routing map can save time when one source appears but another does not.
Check what the Scene panel includes
The Scene panel defines the content Character Animator renders for the live output. Adobe states that only the current scene displayed there is streamed and that content outside the panel is not rendered into the output. This makes the panel a boundary, not a window onto everything visible in the project workspace.
Before you click Stream Live, select the scene you actually intend to broadcast and inspect its framing. If a title, prop or second puppet is beyond the Scene panel edge, it will not become part of the output simply because it remains visible elsewhere in the editing interface. Move the relevant content into the scene frame or adjust the composition before going live.
This boundary is useful when diagnosing a receiver that shows less than expected. If an element is missing, first check whether it is within the current scene and the visible panel area. Then check whether the receiver is cropping or scaling the incoming source. Do not assume the receiver has access to material outside the transmitted scene, and do not build a layout that depends on it.
A second check is scene identity. Projects can contain multiple scenes, and the receiver only needs the output that is currently being sent. Give scenes clear names if you switch between them during a show, and make a deliberate check before changing scenes. If you plan a long-form programme, the same discipline used to prevent unwanted repeats in a YouTube stream playlist applies here: know what is live now and what is meant to follow.
Understand alpha and background behaviour
The Scene panel's background-colour setting is not carried through as a background colour that the receiver inherits. Character Animator's output carries alpha; if the receiving device cannot handle alpha, the background appears black. That difference explains why the scene can look as expected in Character Animator but arrive with a black area or transparency at the other end.
If you want a solid colour or image behind the puppet, make the background part of the composition the receiver is actually given, or add the background in the receiving application. Do not depend on the Scene panel's background-colour setting to be reproduced by the receiver. The decision is about where the final composite is made: Character Animator's outgoing scene, or the broadcast application's composition.
When transparency is important, check the receiver's documentation and inspect its preview. Confirm that the receiving route supports alpha and that the source is interpreted as intended. A transparent area can reveal whatever layer sits below it in a broadcast composition; an unsupported alpha path can instead produce black. Both are plausible symptoms of the same output being handled differently downstream.
For diagnosis, test three simple cases: a puppet over the scene as composed, a receiver with a deliberate background layer, and a receiver with no lower layer. Watch the receiver preview rather than judging only from Character Animator. If the black area disappears when a background layer is added, the issue is likely the receiving composite or alpha handling rather than missing puppet content.
Preview the receiving output
Click Stream Live at the bottom of the Scene panel after confirming the active scene and output preference. Then inspect the receiver's own preview. Check that the puppet is present, the framing is correct, the background behaves as intended and any other broadcast elements are layered in the right order. The receiver's view is the practical test of what your audience will be sent.
Audio should be checked separately. Adobe notes that live output can introduce a small video delay; when audio and video are streamed together, its guide says an audio delay adjustment is usually around 60 milliseconds. Treat that as an initial reference, not a universal setting. The result varies with the computer, output path, receiver and the rest of the broadcast chain.
Record or rehearse a short spoken passage and watch the receiver preview while listening to the intended monitoring path. Look for mouth movement that leads or trails the voice, and adjust audio delay only after identifying which side is early. Adobe notes that delay can be added through external hardware or a streaming application such as OBS Studio or Telestream Wirecast. Check the current documentation for the application and device you use.
The monitoring arrangement matters. Headphones help the performer hear the voice without feeding speaker output back into the microphone, and they make it easier to notice monitoring delay. If the voice sounds doubled, distant or out of time, identify whether the problem is microphone monitoring, application monitoring or the final outgoing mix before changing several settings at once.
Rehearse the entire chain with the actual voice talent and on the intended platform. Adobe recommends testing before performance; that includes the scene, receiver, audio route and final broadcast destination. If you run a longer YouTube show, use a repeatable preflight checklist and assign someone to watch the receiving preview while the performer concentrates on delivery. For a broadcast designed to stay available beyond a single performance, see the practical considerations in running a 24/7 study stream.
When a scene must remain live while your own computer is switched off, StreamNeo can remove the need to keep that computer running by turning an uploaded video into a YouTube live stream. It is a different use from sending a tracked Character Animator performance live: prepare and check the scene output first, then choose the route that fits whether the content is performed live or replayed as a file.
Before committing, compare the operating options on the pricing page. When the file and channel are ready, start free — 24-hour trial, no card.
FAQ
Does Character Animator stream the whole project workspace?
No. It streams the current scene shown in the Scene panel, and content outside that panel is not rendered into the output. Check the active scene and its framing before starting the live output.
Does the receiving application inherit the Scene panel background colour?
No. The output carries alpha, and a receiver that cannot handle alpha may show black instead of a background. Add or select a background in the composition that the receiver actually displays, then verify its preview.
What audio delay should I start with?
Adobe describes an adjustment usually around 60 milliseconds when audio and video are streamed together. Use it only as a starting estimate: rehearse with the actual setup and adjust by listening and watching the receiver output.
Can I use NDI for multiple characters?
Adobe documents NDI as an output path and recommends a wired, professional audio setup and Gigabit Ethernet over Cat 5e or Cat 6 for multi-computer workflows. Confirm the current requirements for your software, devices and network, then rehearse the full route before performing.