Loudness targets
YouTube targets -14 LUFS, broadcast TV -23 LUFS, podcasts -16 LUFS. Check Loudness Radar (Window → Audio Track Mixer → effect slot) to land on target. Mixing without measuring is guessing.
Essential Sound — your mixing assistant
Window → Essential Sound. Tag each clip as Dialogue, Music, SFX or Ambience. Premiere unlocks task-specific tools per category: dialogue gets Reduce Noise and DeReverb; music gets Ducking; ambience gets a clean-up panel.
Dialogue clean-up
Dialogue → Repair → Reduce Noise 4–6, Reduce Rumble 4. Then EQ with Voice preset, lift presence around 3 kHz, gentle cut at 200 Hz. Always A/B with the original — over-processing destroys natural tone.
Music ducking
Tag music as Music, scroll to Ducking, click Generate Keyframes. Premiere reads the dialogue track and automatically lowers the music whenever someone speaks. Manual touch-ups in 30 seconds.
J cuts and L cuts
A J cut starts the audio of the next shot before the video cuts. An L cut keeps the audio of the previous shot running after the cut. Both let dialogue flow naturally across edits — the bedrock of invisible editing.
Audio transitions
Constant Power (default crossfade) preserves perceived volume; Constant Gain causes a midpoint dip. Use Constant Power for music, Constant Gain for very short fades on dialogue.
Sub-mixing
Create a sub-mix bus (Sequence → Add Tracks → Submix). Route all dialogue tracks into one bus, apply compression once, set output level once. The mix becomes far easier to balance.
Final mix check
Listen on three systems: studio monitors, laptop speakers, phone earbuds. A mix that works everywhere is the only mix that's actually finished.