In this video I show you how I duck vocals to music for live stream or broadcast—the kind of “radio station” effect where the music drops smoothly the moment you talk, then comes back up when you stop. I start by setting proper input levels from my sound card. My rule is simple: keep the vocal level sitting in the green (not pushing into yellow), because if your gain staging is wrong, the ducking will never feel clean. From there I build a basic vocal chain: gate, EQ, then compression. The gate helps me deal with fan/room noise, and I’m careful with the release so it closes smoothly when I stop talking (not too sharp, not too slow). With EQ, I don’t want you to “cram” my settings—your voice texture matters—but I show how I cut unnecessary high end for spoken vocals, tame some harsh/sibilant areas, and add a bit of thickness. Then I use a mild compressor (I usually like a 3:1 style approach) aiming around ~3 dB reduction to keep the voice steady. Finally, the main trick: I drop the music level a bit under the vocal, insert a compressor on the music, and sidechain it to the vocal. I explain how I set ratio/threshold and especially release so the music ducks enough (around -4 to -5 dB reduction for my example) without staying down too long. No rigid rules—mess around and tune it to what you want to hear.