Tools

Stream In

Capture any browser tab as a CV-modulatable, filterable audio source inside the rack.

What is a Stream In?

A virtual modular synth that lives in a browser sits next to every other tab on your computer. The browser knows how to capture audio from any of those tabs - the YouTube tab playing a song, the Spotify tab streaming a podcast, a SoundCloud tab loaded with a friend's demo - and turn that captured audio into a signal a webpage can read. A tab capture module wires that capability into the rack: any other tab becomes a Eurorack-level audio source, ready to be patched into a filter, a delay, or anything else.

This is more than a convenience. It collapses the wall between the rack and the rest of your computer. You can filter sweep an album, granularize a podcast, layer a hardware synth (running through a DAW in another tab) on top of a Webrack VCO, or use a backing track as the carrier for a vocoder-style chain. The browser's strict tab isolation means you cannot accidentally route audio between tabs without permission - but with permission, the entire web becomes a sample library.

The mechanism is the Screen Capture API (getDisplayMedia), originally built for screen-sharing tools like Google Meet and Zoom. Picking Tab instead of Window or Entire Screen and ticking Share tab audio gives the page a live MediaStream of just that tab's audio - no video required for our purposes, but the API delivers video too which we use for a small live preview.

Our Stream In

Webrack's Stream In is 10 HP and uses getDisplayMedia with tab capture to pull audio from any other browser tab. Click the on-panel Capture button, the OS's tab picker opens, you choose a tab and tick Share tab audio, and the L/R/Mono outputs come alive. The captured tab's video is shown as a small live preview on the panel so you know which tab you connected to.

We use a separate worklet input from Audio In, which means microphone capture and tab capture can run *simultaneously* - sing into a mic while a backing track plays from a tab, both feeding the same Mixer and the same Output. Latency is the usual one-block, roughly 3ms.

Tab audio capture is a Chromium-only browser API, so this module only works in Chrome, Edge, Brave, Arc, and Opera. Firefox and Safari users will see a friendly error - in those browsers, the workaround is to run the source app elsewhere and route via Audio In from a virtual cable like BlackHole or Soundflower.

In a patch

The simplest creative move is a filter sweep on captured audio. Patch Stream In L (or M) into a Filter audio-in, run an LFO slowly into the filter's cutoff-cv, and listen as the LFO scrubs through the spectrum of a song. Boost resonance and the filter rings out individual frequencies - melody hidden inside the spectrum surfaces, then disappears.

Drown captured audio in a long Reverb for ambient remixes - a podcast becomes a cathedral drone, a pop track becomes a smear of harmonic content. Combine with a Delay for dub-style tape echoes. Or feed a Granular processor for grain-cloud reconstructions of any input audio.

Use it as a sidechain or analysis source. Patch the captured audio into a Spectrum Analyzer to see what frequencies are present in your reference track. Run it through an envelope follower (a Filter slow-tracking the level) to extract a CV that ducks a VCA rhythmically against the source.

Outputs

L (audio) — Left channel from the captured tab, at Eurorack +/-5V. Silent until you click Capture and select a source.
R (audio) — Right channel. Stereo content from the tab is preserved.
M (audio) — Sum (L+R)/2. Use when downstream effects need a single channel.

Controls

Gain — Input gain, 0-2x. Many tabs play loud - reduce gain if downstream modules clip.
Mute — Silences the output without ending the capture session. Re-enable instantly without re-prompting for the tab.

← Back to all modules