top of page

The Master Guide / 13 Sound design and mix · The mix

Planned silence and playback compression

A silence the treatment plans drops to air or room tone, never to digital zero. Check that each one still reads after a platform squashes the dynamics.

A silence the treatment plans drops to air or room tone, never to digital zero. Check that each one still reads after a platform squashes the dynamics.

Planned silences come from the treatment's contrast map (see Writing the treatment → The sound sheet). Mark each in the timeline as one of two kinds:

  • air: the sound drops to room tone or air;

  • zero: a full cut-away to nothing, only where the treatment writes one, declared as such.

At delivery, a digital zero outside a declared full cut-away fails the delivery check, and every quiet stretch is listed for review (see Graphics, captions, versions and delivery → Exports and the delivery check).

Playback compression. YouTube's Stable Volume starts switched on for its Android and TV apps, and Meta levels clips on the phone; both squash dynamics. So run each planned quiet moment, with 2 s either side, through a compression stand-in, and listen: do its words and its key sound still read? It is a listen flag, never a target [proven] as mixing craft. Level the whole mix first, then keep the moment with its 2 s either side, so the leveller hears what comes before the moment (steps below).

Do it yourself

Run a quiet moment through the compression stand-in (ffmpeg)

  1. Run the whole mix through the leveller and keep the moment:

ffmpeg -i mix.wav -af "dynaudnorm=f=200:g=15:p=0.95:m=10:s=0,atrim=19.0:25.5,asetpts=PTS-STARTPTS" Q1_00m21.0s_compressed.wav

  1. It levels the whole mix first, then keeps 19.0–25.5 s (a quiet moment at 21.0–23.5 s, with 2 s either side), so the leveller hears what comes before the moment.

  2. Listen to the file: do its words and its key sound still read?

Sources

bottom of page