Capture a Virtual Audio Input

Bring audio from any Windows application into Backstage by capturing a render endpoint. The application sends its audio to a virtual audio device; Backstage captures that device and feeds it into the mixer as a live stream. No virtual monitor is needed — this asset is independent of video capture.

Before you start

• A Brainsalt virtual audio device is installed and visible in Windows Sound settings (listed as Backstage Virtual Audio or similar).
• The application whose audio you want to capture is configured to output to that virtual device. Set this in Windows Settings > Sound > Volume mixer (per-app output).
• A project is open in Backstage with at least one audio channel available in the Sequencer.

Steps

1. Add a Virtual Audio Input

In the Asset Manager, expand the Audio section and click the + button. Select Add Virtual Audio Input from the popup. A new asset appears in the Audio list.

2. Choose the endpoint to capture

Select the new asset. In the Inspector, open the Virtual Audio Input section. Use the Endpoint dropdown to pick the audio device you want to capture. Virtual audio devices are listed first. The dropdown refreshes live, so a device added while Backstage is running appears without a restart.

3. Place the asset on a channel

Drag the asset from the Audio list onto an audio channel in the Sequencer. Backstage creates a looping event — the capture runs for as long as the transport is playing, just like a live video input.

4. Verify the capture

Start playback. Below the Endpoint dropdown the Inspector shows a status line. When capture is active it reads Capturing: followed by the resolved device name. If the endpoint cannot be found it reads Not capturing - retrying followed by the name that was requested.

Common Mistakes

• "Not capturing" after choosing an endpoint. The endpoint name is matched as a case-insensitive substring. If nothing matches, capture stops and the log lists every active endpoint it did find. Check for a typo, or open the dropdown again and pick the device from the list.

• No audio despite "Capturing" status. The application may still be sending audio to a different device. Confirm in Windows Settings > Sound > Volume mixer that the application's output is pointed at the same device Backstage is capturing.

• Capturing your own output by mistake. Any render endpoint can be captured, including real hardware outputs. If you capture an endpoint that Backstage itself plays to, the mix feeds back into itself. Use a dedicated virtual device to avoid this.

• Audio sounds wrong after changing the device's speaker configuration. The stream is built at the endpoint's actual sample rate and channel count. If the configuration changes (e.g. from 5.1 to stereo), Backstage detects the mismatch on the next project load and re-formats the stream automatically. Saving and reloading the project resolves it.

Tips

• You can capture several endpoints independently by adding multiple Virtual Audio Input assets, each pointed at a different device.

• If the endpoint briefly drops out (for example after a driver update), capture retries automatically and recovers on its own. The Inspector shows the retry status.

• The virtual audio device does not need to be the Windows default output. Point individual applications at it in Windows per-app output settings and leave the rest of the system on real speakers.

• Multichannel endpoints (up to 7.1) are captured at their native channel count — Backstage does not downmix to stereo.

• This asset is audio-only and does not require a virtual monitor. To capture both video and audio from an application, add a Virtual Audio Input alongside a separate Capture a Virtual Monitor asset.

Related

Capture a Virtual Monitor — capture an application's video output via a virtual monitor.

Import Video & Audio Assets — add pre-recorded audio and video files to the project.

Create a Sequence and Add Content — build a sequence and place assets on channels.