Problem
Dialogue recorded at different gain levels requires clip-by-clip normalization before a consistent mix is possible.
Solution
Root Cause / Diagnostic:
Recording dialogue across varying scenes or distances without consistent preamp gain staging results in wide clip-to-clip RMS disparities. When these clips are edited back-to-back, listeners experience abrupt jumps in vocal presence and noise floor, making a unified mix impossible. Relying on manual fader riding without baseline normalization creates erratic gain staging into downstream compressors and limiters.
Actionable Fix:
1. Select all dialogue clips on the timeline and apply batch loudness normalization to -23 LUFS (integrated) or normalize peak amplitude to -3 dBFS.
2. Group normalized dialogue onto dedicated dialogue buses and insert a gentle leveling compressor (2:1 ratio, 30ms attack, 100ms release) to smooth remaining micro-dynamics.
3. Monitor the dialogue bus using a dedicated ITU-R BS.1770 loudness meter to confirm integrated dialogue levels sit consistently between -16 and -14 LUFS for web delivery.
Pro Tip:
Always normalize dialogue clips to target LUFS loudness rather than peak dBFS, as perceived loudness correlates with integrated acoustic energy rather than momentary transient spikes.
Recording dialogue across varying scenes or distances without consistent preamp gain staging results in wide clip-to-clip RMS disparities. When these clips are edited back-to-back, listeners experience abrupt jumps in vocal presence and noise floor, making a unified mix impossible. Relying on manual fader riding without baseline normalization creates erratic gain staging into downstream compressors and limiters.
Actionable Fix:
1. Select all dialogue clips on the timeline and apply batch loudness normalization to -23 LUFS (integrated) or normalize peak amplitude to -3 dBFS.
2. Group normalized dialogue onto dedicated dialogue buses and insert a gentle leveling compressor (2:1 ratio, 30ms attack, 100ms release) to smooth remaining micro-dynamics.
3. Monitor the dialogue bus using a dedicated ITU-R BS.1770 loudness meter to confirm integrated dialogue levels sit consistently between -16 and -14 LUFS for web delivery.
Pro Tip:
Always normalize dialogue clips to target LUFS loudness rather than peak dBFS, as perceived loudness correlates with integrated acoustic energy rather than momentary transient spikes.