The Master Guide / 13 Sound design and mix
Sound design and mix
A film's sound is built in three layers, real sound first, against the locked cut and with the music playing. Both come before this chapter: the cut in Editing, the music in Music. This chapter covers where each sound comes from, the sound effects, the mix, the placed voice, loudness and the stems.
A film's sound is built in three layers, real sound first, against the locked cut and with the music playing. Both come before this chapter: the cut in Editing, the music in Music. This chapter covers where each sound comes from, the sound effects, the mix, the placed voice, loudness and the stems. The voices are recorded in Voice and Arabic dialects. The sound you write into each shot's Seedance prompt, its AUDIO block, is a different subject: it is taught in the Seedance prompt topic Video with Seedance → Sound inside the shot.
Where sound comes from
The three layers of sound, the source order, what to keep or mute of each shot's own sound, and reusing sound from earlier takes.
Three layers and the order of the sound work: Every sound in the film comes from one of three layers, and each layer has one source.
Real sound first: the source order: For each row of the sound list, take the first source that can give the sound, and write down why.
Keeping or muting in-shot sound: The shot's own sound stays where it works.
Reusing sound from earlier takes: Your earlier takes are a free sound library recorded in the right rooms.
Effects and beds
The effects list, real CC0 recordings, designed effects, where each effect sits, and the earshot check.
The SFX list and its Lead column: One row per effect, written from the treatment's sound idea and the locked cut, after the music is chosen.
Real recordings from Openverse (CC0): Openverse is a free, open search of openly licensed audio.
Designed effects with ElevenLabs: Design only what reality can't give.
Placing effects: Every effect lands on its picture event, stays off the words, and supports whatever leads the moment.
The earshot listening check: Earshot is a listening aid that reports and never decides; your ear decides.
The mix
Mixing around the words, the placed voice, checks on small speakers, planned silence, loudness targets, stems, and the five moments to check by ear.
The mix with speech as the anchor: Mix the words first and build everything else around them [official guidance].
The placed voice: A studio-dry voice in a wide street shot is the clearest tell of artificial sound.
Checking the mix on small speakers: Before you approve a mix, measure whether the words survive on small speakers and in mono.
Planned silence and playback compression: A silence the treatment plans drops to air or room tone, never to digital zero.
Loudness: −23 TV, −16 online, −14 short social: Hit each delivery's loudness target, and measure it on every file.
Stems: When the deliverables ask for stems, deliver one file per part of the mix that adds up exactly to the mix.
The five moments to check by ear: After the mix, listen again at the five places where a mistake is most likely, or would cost the film most (decided).
