7 min read
Sound design for film: building a scene from separate layers
Almost nothing you hear in a finished film was recorded as one sound. Here is how the layers are built and kept manageable.
Sound design is the craft of inventing what a scene sounds like when reality is either unavailable or unconvincing. A door in a thriller is rarely a door: it is a hinge, a body impact, a low sub and a room tail, aligned to the frame where the handle turns.
The layer method
- Attack: the transient that sells the moment, usually a close recorded impact.
- Body: the sustained part that gives the object mass and material.
- Tail: the decay, often reverb or a recorded space, that places the event in a room.
- Sweetener: a subtle extra, sub bass or a metal ring, that makes it feel bigger than life.
Why source quality decides everything
Design work stretches, pitches and reverses material far beyond normal playback. A recording made at 96 kHz and 24 bit survives being pitched down two octaves; a 128 kbps MP3 falls apart into swirling artefacts. Record and store sources at the highest rate you can, and only convert on delivery.
Field recordings are the raw material of the trade. Split multichannel takes into mono stems so each microphone becomes an independent layer you can pitch and place separately.
Keeping a library you can actually search
A design session lives or dies on how fast you find the right file. Adopt a consistent naming convention, ideally UCS, write descriptions into the metadata rather than the folder structure, and batch rename new recordings the day you make them.
Delivering clean stems
The mix needs separated stems: dialogue, music, effects, foley and ambience. Design layers get printed into the effects stem at the sample rate and bit depth of the project, usually 48 kHz and 24 bit. Check every export for clipping and for accidental channel duplication before it leaves your machine.