Setting Up Virtual Backgrounds and Scenes in vMix

The short answer

vMix is Windows-based live production software. It replaces backgrounds by keying out a uniform backdrop, usually a green screen, from a camera input and compositing the person over a still image, a video or a virtual set. This is a video operation only: it does not change or improve the audio from a webcam microphone. Your speech capture depends on which audio input you select in vMix, how it is routed to the conferencing app or stream, and whether audio stays in sync with the processed video.

One condition limits this answer. No confirmed product evidence was supplied for this guide, so it gives no edition-specific feature lists, no count of bundled virtual sets, no prices and no hardware minimums. Check the current vMix documentation for your edition before relying on any particular feature.

How background replacement differs from a simple virtual background

Conferencing apps often offer AI background blur or replacement that guesses where the person ends and the room begins. vMix is generally used with chroma keying instead, which removes a specific colour range. With even lighting and a clean backdrop, keying usually gives sharper edges than segmentation. The cost is that it needs a physical backdrop and careful lighting. A virtual set goes a step further by placing the keyed person inside a designed scene rather than in front of a flat image.

Typical setup workflow

• 1. Add the webcam or camera as an input in vMix.
• 2. Light the backdrop evenly and keep the speaker a short distance in front of it to reduce colour spill and shadows.
• 3. Enable colour keying on the camera input and adjust the key until the backdrop disappears without eating into hair or clothing edges.
• 4. Add the background image, video or virtual set and layer the keyed camera over it.
• 5. Save the combination as a scene or preset so it can be recalled during a live session.
• 6. Send the result to the stream, recording or a conferencing app through the vMix output you use.

What this means for speech capture

In vMix, a webcam’s built-in microphone typically shows up as its own audio source, separate from the video. Selecting the camera picture does not guarantee that the matching microphone is the one feeding your output. Confirm in the audio mixer which source is live and which ones are muted.

• Pick the intended microphone. If a USB or headset microphone is also connected, decide which one carries speech and mute the rest to avoid echo and doubled voices.
• Watch lip sync. Keying and compositing can add video processing delay. If speech arrives before the lips move, look for a per-input audio delay setting and adjust it by ear and eye.
• Avoid double processing. If vMix audio goes into a conferencing app, that app may apply its own noise suppression and echo cancellation on top of any processing in vMix. Use one stage of processing where possible.
• Select matching devices in the conferencing app. When video goes out through a virtual camera, choose the corresponding audio route in the app as well. Otherwise the app may still pull the raw webcam microphone directly.

A green screen does not help a webcam microphone. Hard, empty walls near a backdrop can add room reflections, which make speech sound distant. Soft furnishings nearby, or a microphone placed closer to the mouth, address this more directly than any video setting.

System requirements and edition features: gaps to check

The materials for this guide contained no confirmed system requirements, edition comparison or virtual set count, including the claim of six virtual sets in a Pro edition. As a general principle, real-time keying and compositing load the graphics processor, and running several inputs at once raises the demand. For exact processor, graphics card and operating system requirements, and for which editions include virtual sets, refer to the official vMix specifications for the version you plan to install.