Been using Murf for onboarding videos. The voiceovers are clear, but when I add background music from the library, the voice gets lost. Even lowering the music track feels like a compromise—it either disappears or still interferes.
What’s the proper technique here? Are there specific volume levels or audio profiles that work best? Is it about choosing a different type of music entirely, or does Murf have built-in ducking features I'm missing?
Music choice makes a bigger difference than you'd think. I've had good luck with ambient or "corporate chill" tracks that have minimal mid-range frequency. The voice sits in those mids, so you want music that stays out of that lane.
Since Murf lacks built-in ducking, you can fake it a bit in their editor. Lower the music track to about -20dB, then add short volume dips at the start of key voice phrases. It's manual, but it works for short clips without jumping to a DAW.
What genre are you using from their library? Some are just too busy for clear speech.
Pipeline Pilot
Murf's built-in editor is quite limited for this. You're right that simply lowering the track is a compromise; it reduces overall energy but doesn't solve the masking of speech transients.
The core issue is frequency masking, not just volume. Even a quiet track with prominent energy in the 300 Hz - 3 kHz range will obscure the voice. You need to use an equalizer, either within Murf if it has one or by exporting the tracks for external processing. Apply a broad cut in that midrange to the music track, around -6 dB with a moderate Q. This creates a spectral "pocket" for the voice to sit in. Start with the music at -18 dB LUFS relative to the voice at -3 dB LUFS, then adjust the EQ before touching the fader again.