Effects
Audio effects
Every audio clip — and every video clip's embedded track — carries keyframeable gain, fades, a parametric EQ and independent pitch control.
Gain & fades
gain is a linear multiplier from 0 to 4 (up to +12 dB), keyframeable via the gain path. Every clip also carries a universal audioMix.volume trim on top of it. Drag the fade handles at either edge for a fadeIn/ fadeOut ramp, each with its own duration and easing curve. Fades multiplythe gain envelope rather than overriding it — DaVinci Resolve's model — so a fade and gain keyframes always compose instead of fighting each other.
Parametric EQ
eq is an unbounded list of bands, not a fixed low/mid/high triple — add as many as the mix needs. Each band is an RBJ biquad filter. A clip with eq: nullhas EQ bypassed entirely (no filter nodes in the chain); an empty list is EQ-on but flat. Each band's gain, frequency and Q are independently keyframeable via eq.<bandId>.gain, .freq and .q.
| Field | Range | Notes |
|---|---|---|
type | — | lowshelf, peaking, highshelf, lowpass, highpass, or notch. |
freqHz | 20 – 20000 | Band center/corner frequency. |
gainDb | −24 – 24 | Boost/cut for shelf and peaking bands; ignored by pass and notch types. |
q | 0.05 – 18 | Bandwidth — higher is narrower. |
Time-stretch & pitch
playback.speed retimes a source-backed clip. keepPitch (on by default) holds pitch steady while speed changes; turn it off and pitch shifts with speed, tape-style. pitchSemitones is an independent shift from −24 to 24, keyframeable via the pitch path, layered on top of whatever keepPitchis doing. A real pitch shift can't run live off a plain media element, so it renders through an offline audio bake once and caches the result. Video clips carry this exact same keepPitch/pitchSemitones contract for their embedded audio track.
Voice role
role tags a clip as voice, music, sfx or none — auto-detected on import from the filename, always overridable by hand. Setting a clip to voice forces keepPitch on and surfaces vocal-tuned EQ presets.
This is a different, separate sound system from typing sound: text layers have their own per-word/letter typing sound — see Text animation. For where clips live and how tracks are organized, see The timeline.