Trim / AudioTrim

Set of filters to trim audio and video:

  • Trim can trim video only clips and also clips with both audio and video.

  • AudioTrim can trim audio only clips and also clips with both audio and video.

Trim

Trims a clip – removes frames from one or both ends.

Trim cannot trim a clip if there is no video. Use AudioTrim for that case.

The resulting clip starts with frame number 0, and this must be taken into account when making additional edits using that clip. To view a clip's frame number at any point in your script, temporarily insert an Info or ShowFrameNumber statement.

Syntax and Parameters

For convenience, Trim can be called in four slightly different ways:

Trim (clip, int first_frame, int last_frame , bool "pad", bool "cache")
clip

Source clip; all color formats and audio sample types supported.

first_frame, last_frame

Returns a clip starting at first_frame and running up to and including last_frame.

  • For example, Trim(3, 5) returns source clip frames 3, 4 and 5.

  • Remember, AviSynth starts counting at frame 0.

  • If you set last_frame to 0, you will get a clip starting at first_frame and running to the end of the clip.

pad

pad causes the audio stream to be padded to align with the video stream. Otherwise the tail of a short audio stream is left so. You should use pad=false when the soundtracks being joined were originally contiguous – compare to UnalignedSplice.

Default: true

cache

When set to false: lower memory consumption but may be slower. Benefits heavily depend on how trimmed clips are used later. See issue #274 for more information.

Default: true


Trim (clip, int first_frame, int -num_frames , bool "pad", bool "cache")
first_frame, -num_frames

With a negative value for the second argument, you get a clip starting at first_frame and running for num_frames frames.

  • For example, Trim(0, -4) returns source clip frames 0, 1, 2 and 3.


Trim (clip, int first_frame, int "end" , bool "pad", bool "cache")
first_frame, end

Returns a clip starting at first_frame and running up to and including frame end.

  • For example, Trim(3, end=7) is equivalent to Trim(3, 7); both return frames 3, 4, 5, 6 and 7.

  • end default = 0; must be >= first_frame.

  • Unlike the last_frame syntax, the end syntax has no discontinuous boundary values: end=0 means end at frame 0. This feature is useful in avoiding unexpected boundary conditions in your user functions.


Trim (clip, int first_frame, int "length" , bool "pad", bool "cache")
first_frame, length

Returns a clip starting at first_frame and running for length frames.

  • For example, Trim(3, length=5) is equivalent to Trim(3, -5); both return frames 3, 4, 5, 6 and 7.

  • length default = 0; must be >= 0.

  • Unlike the num_frames syntax, the length syntax has no discontinuous boundary values: length=0 means return a zero length clip. This feature is useful in avoiding unexpected boundary conditions in your user functions.

AudioTrim

AudioTrim trims a clip based on time, not on frames. This is most useful for audio-only clips, where "frames" have no meaning anyway, and you may want to edit with finer precision than whole frames (at 30fps, 1 frame=33.3ms).

All time arguments are in seconds, floating-point.

  • Trims on audio-only clips are accurate to the nearest audio sample.

  • Trims on clips with video are accurate to the nearest whole video frame.

AudioTrim cannot trim a clip if there is no audio. Use Trim for that case.

The resulting clip starts with time = 0.0, and this must be taken into account when making additional edits to that clip. To view a clip's time at any point in your script, temporarily insert an Info or ShowTime statement.

Syntax and Parameters

For convenience, AudioTrim can be called in four slightly different ways:

AudioTrim (clip, float start_time, float end_time, bool "cache")
clip

Source clip; all color formats and audio sample types supported.

start_time, end_time

Returns a clip starting at start_time and running up to and including time end_time.

  • For example, AudioTrim(3.0, 5.0) returns source clip from time 00:00:03.000 to 00:00:05.000.

  • If you set end_time to 0.0, you will get a clip starting at start_time seconds and running to the end of the clip.

cache

When set to false: lower memory consumption but may be slower. Benefits heavily depend on how trimmed clips are used later. See issue #274 for more information.

Default: true


AudioTrim (clip, float start_time, float -duration, bool "cache")
start_time, -duration

With a negative value for the second argument, you will get a clip starting at start_time and running for duration seconds.

  • For example, AudioTrim(0.0, -4.0) returns the source clip from time 00:00:00.000 to 00:00:04.000.


AudioTrim (clip, float start_time, float "end", bool "cache")
start_time, end

Returns a clip starting at start_time and running up to and including time end.

  • For example, AudioTrim(3.0, end=7.0) is equivalent to AudioTrim(3.0, 7.0)

  • end default = 0.0; must be >= start_time.

  • Unlike the end_time syntax, the end syntax has no discontinuous boundary values: end=0.0 means return a zero-length clip. This feature is useful in avoiding unexpected boundary conditions in your user functions.


AudioTrim (clip, float start_time, float "length", bool "cache")
start_time, length

Returns a clip starting at start_time and running for length seconds.

  • For example, AudioTrim(3.0, length=4.0) is equivalent to AudioTrim(3.0, -4.0)

  • length default = 0.0; must be >= 0.

  • Unlike the duration syntax, the length syntax has no discontinuous boundary values: length=0.0 means return a zero-length clip. This feature is useful in avoiding unexpected boundary conditions in your user functions.

Examples

  • Return only the first frame (frame 0):

    Trim(0, -1)
    Trim(0, end=0)
    Trim(0, length=1)
    
  • Return only the last frame:

    Trim(FrameCount, -1)
    Trim(FrameCount, end=Framecount)
    Trim(FrameCount, length=1)
    
  • Return frames 100-199 (duration=100):

    Trim(100, 199)
    Trim(100, -100)
    
  • Delete the first 100 frames; audio padded or trimmed to match the video length.

    Trim(100, 0) # (pad=true by default)
    
  • Delete the first 100 frames of audio and video; the resulting stream lengths remain independent.

    Trim(100, 0, pad=false)
    
  • Trim audio if longer than video (pad does affect this action):

    Trim(0, FrameCount-1)
    
  • Audio will be trimmed if longer, but not padded if shorter to frame 199:

    Trim(100, 199, pad=false)
    
  • AudioTrim: trim video if longer than audio:

    AudioTrim(0, AudioDuration)
    
  • AudioTrim: keep the audio between 1.0 and 6.5 seconds (ie, delete the first second, keep the following 5.5 seconds):

    AudioTrim(1, 6.5)
    AudioTrim(1, -5.5)
    AudioTrim(1, length=5.5)
    

Changelog

Version

Changes

AviSynth+ 3.7.2

Added cache parameter.

AviSynth 2.6.0

Added AudioTrim.
Added explicit length and end parameters.

AviSynth 2.5.6

Added pad audio parameter.

$Date: 2022/03/12 14:59:42 $