Audio

Window ▸ VSK Audio Mixer - a full show mixer, field-tested by ear on real hardware (SDI camera mic + a Focusrite Scarlett 4i4 WASAPI interface, August 2026): strips, channel routing, gain/pan/delay, meters, recording with audio and background audio are all operator-confirmed. What still awaits a studio day is listed at the bottom - nothing there reads as shipped.

Strips

• SDI:N - each camera’s embedded audio (stereo - a UE/Blackmagic limit; 16-channel embed isn’t available).
• Cap:<device> - up to 4 WASAPI capture strips for audio interfaces / mics. No ASIO - switch the interface to its WASAPI mode.

Per-strip controls

Editable label · Gain (−24…+24 dB, a real trim - positive gain genuinely boosts, and a hot source can clip near +24) · Delay (0-2000 ms, lip-sync) · Ch channel routing · Pan · vertical fader beside split L/R meters · MUTE · SOLO · AFV (audio-follows-video - ~150 ms crossfade on camera cuts) · 3-band EQ (100 Hz / 1 kHz / 8 kHz, ±12 dB) · Gate · Compressor. All DSP defaults off = bit-neutral passthrough. Every value has a tooltip and a ↺ reset-to-default arrow (shown only while the value differs from default - the Details-panel convention).

Channel routing (Ch) - the mono-mic fix

The Ch dropdown routes the strip’s input channels before Pan: Stereo (unchanged) · L → both · R → both · Mono sum ((L+R)/2 on both).

A mono camera mic usually arrives on just one SDI channel (often ch 2 = right) - you hear it in one ear and the meters show one live bar. Set Ch to spread it to both channels. Pan cannot do this: Pan is a stereo balance (it only attenuates left or right), so panning a right-channel-only mic left just fades it to silence. Route with Ch first, then use Pan for placement.

Meters - final result, split L/R

Every strip and the master show two vertical bars (green / −12 dB yellow / −3 dB red) with L / R labels and whole-dB readouts underneath; the master adds a numeric dB label. They meter the final processed signal - Ch routing, Pan, EQ/Gate/Comp, Gain and the fader all included - i.e. exactly what the Program bus receives. A stereo source moves both bars; a mono-on-one-channel mic visibly moves one.

Master & monitor - the isolation rule

The MASTER card has the show fader, split meters, mute, a 1 kHz test tone and the Monitor Output device picker. Monitoring uses the Windows app-session volume on the chosen endpoint - applied by the OS after the engine - so the operator’s headphone level can never change what goes to air. The master fader is on-air: it shapes recordings and the SDI embed too.

The mixer also enables the editor’s “Allow background audio” automatically - minimizing or unfocusing the Unreal window never mutes a running show (the editor’s default is mute-in- background, which is unacceptable on a broadcast machine).

Signal flow

strip (SDI:N / Cap:X)
  → VS_Submix_<Strip>   (delay · gain · EQ/gate/comp · Ch routing · pan)
    → VS_Submix_Program  (master fader)
      → engine master → SDI embed (stereo 48 kHz) + recorder tap
operator ears: Windows session volume on the Monitor endpoint (post-engine)

Every control is also a bus channel (Audio.<Strip>.Volume, Audio.Master.Mute, …) for web boards, Stream Deck and key bindings - intentionally not listed in the VSK Control panels (the mixer tab is their home).

First run

Press Create SDI Input Assets once (it persists bCaptureAudio on the camera sources), then Play. Opening the mixer tab creates/repairs the /Game/VSK/Audio submix chain automatically - the DSP chain latches at Play start, so restart Play once after the first open.

Still in studio verification

Built end to end but not yet field-verified: the SDI program-output audio embed (needs a camera + SDI-monitor day) · AFV on real camera cuts · the web audio board module. Backlog beyond that: Video-Hub file/asset audio · surround / channel-pair selection · sub-buses · DSP params on the web · NDI audio embed.