Voiceover & Music Mixing Masterclass: EQ Pocketing, Auto-Ducking & -14 LUFS Standards

1. Why Poor Audio Mixing Kills Viewer Retention

In digital video production, viewers will tolerate 720p or slightly grainy footage, but they will instantly swipe away or close a tab if the audio is muddy, deafening, or the voiceover is drowned out by loud background music. Psychoacoustic studies demonstrate that subconscious listening fatigue is the primary driver of mid-video drop-offs.

2. Frequency Carving: EQ Pocketing for Vocals

The human speaking voice fundamentally lives in the frequency spectrum between 300 Hz and 3,500 Hz, with sibilance and clarity extending up to 6 kHz. When a background music track features synthesizers, guitars, or pianos occupying those exact same frequencies, destructive phase and frequency masking occurs.

• Music Track High-Pass Filter: Roll off sub-bass rumble below 35–40 Hz using a clean 18 dB/octave low cut.
• Vocal Pocket Dip: Apply a subtle bell parametric EQ cut on the music track at 1.5 kHz – 2.8 kHz by -2.5 dB to -4 dB with a Q factor of 1.2. This immediately "clears a path" for your vocal track without dulling the overall song.
• Gentle High Shelf: If cymbals clash with spoken consonants ("S", "T", "Ch"), attenuate above 10 kHz on the music by 1.5 dB.

3. Transparent Sidechain Auto-Ducking

Rather than manually keyframing volume faders across a 20-minute video, leverage automated sidechain ducking:

1. In Premiere Pro: Open the Essential Sound panel, tag your voice track as Dialogue, tag your music as Music, and check Ducking against dialogue. Set ducking amount to -15 dB with a 400ms fade speed.
2. In DaVinci Resolve Fairlight: Insert the Fairlight Dynamics processor on the music track. Enable Sidechain Listen (Track In), route the dialogue track to the Sidechain Bus, and set Ratio to 3.5:1 with an Attack of 45ms and Release of 350ms.
3. In CapCut: Select the audio clip, toggle on Audio > Voice Isolation > Auto-Ducking to keep volume automatically reduced while dialogue is detected.

4. Loudness Normalization: The -14 LUFS Standard

YouTube, Spotify, and TikTok automatically apply loudness normalization algorithms to prevent listener volume shocks: