Skip to content

Commands

clip.remove_silences

Cuts the silent stretches out of a clip, leaving one clip per stretch of speech.

Last updated:

Replaces the clip by one clip per non-silent stretch of its source range. Silences shorter than min_silence are kept; padding of speech is kept on each side of a cut. With ripple track the pieces close up and the later items of the track follow. The silences come from the analysis of the asset and are supplied by the service in resolved; the command never starts a job.

Arguments#

ArgumentTypeRequiredDescription
item_idstringyesA clip of a video asset with audio.
min_silencestringnoA length; shorter silences are ignored. Default "0.4s".
new_idsarray of stringnoIds for the second, third, ... pieces, in order.
paddingstringnoA length, may be 0; speech kept around each cut. Default "0.1s".
resolvedobjectnoFilled by the service from the silence analysis of the asset; never sent by a caller.
rippletrack, nonenotrack (default) or none.
threshold_dbintegernoNoise level of silence: -30, -35 (default), -40 or -45.

Example#

As one operation of an edit batch (the edit tool of MCP, vukem-video edit --ops, or POST /api/v1/projects/{id}/edit):

JSON
{
  "op": "clip.remove_silences",
  "args": {
    "item_id": "itm_m1n2p3q4r5",
    "min_silence": "0.4s",
    "padding": "0.1s",
    "ripple": "track"
  }
}

Errors#

E_VALIDATION, E_INVALID_TIME, E_NOT_FOUND, E_TYPE_MISMATCH, E_ID_EXISTS, E_INVALID_STATE, E_LIMIT

Scope and phase#

Scope edit, phase alpha. Run vukem-video command:help clip.remove_silences for the same description from the binary.