Build a vertical social clip

Turn a landscape source into a 9:16 clip for TikTok, Reels, or Shorts: pick a 15-second range, standardize the frame rate, convert to vertical, fade in, and lay music over it.

The plan

The cut is the segment's start/end; everything else is an operation. Scale to the target height first, then center-crop the width — that fills the frame instead of letterboxing it.

from videopython.editing import VideoEdit, SegmentConfig, Resize, Crop, ResampleFPS, Fade

edit = VideoEdit(segments=[SegmentConfig(
    source="raw_footage.mp4",
    start=30.0,
    end=45.0,
    operations=[
        ResampleFPS(fps=30),
        Resize(height=1920),                       # scale to height, keep aspect
        Crop(width=1080, height=1920, mode="center"),
        Fade(mode="in", duration=0.5),
    ],
)])

edit.validate()
edit.run_to_file("social_clip.mp4")

Add a music bed

Audio mixing happens on a Video, not in the plan, so render first and mix after:

from videopython.base import Video

(Video.from_path("social_clip.mp4")
      .add_audio_from_file("upbeat_music.mp3")     # overlay=False replaces instead
      .save("social_clip.mp4"))

To duck or mute the source audio inside the plan instead, add VolumeAdjust — it is an audio-only effect and takes a window:

from videopython.editing import VolumeAdjust, TimeRange

VolumeAdjust(volume=0.2)                                   # whole segment
VolumeAdjust(volume=0.0, window=TimeRange(stop=2.0))       # mute the first 2s

The same plan as data

For a UI or an LLM that stores plans, use the dict (JSON) form — same models, same validation:

from videopython.editing import VideoEdit

plan = {
    "segments": [{
        "source": "raw_footage.mp4",
        "start": 30.0,
        "end": 45.0,
        "operations": [
            {"op": "resample_fps", "fps": 30},
            {"op": "resize", "height": 1920},
            {"op": "crop", "width": 1080, "height": 1920, "mode": "center"},
            {"op": "fade", "mode": "in", "duration": 0.5},
        ],
    }],
}

edit = VideoEdit.from_dict(plan)
edit.validate()
edit.run_to_file("social_clip.mp4")

Reframe around a speaker instead of center-cropping

A center crop cuts the subject in half when they stand off-axis. With the [ai] extra, FaceTrackingCrop follows them:

from videopython.ai import FaceTrackingCrop

operations = [
    FaceTrackingCrop(target_aspect=(9, 16), framing_rule="headroom", max_speed=0.1),
    Fade(mode="in", duration=0.5),
]

max_speed bounds how fast the virtual camera may move, which keeps the result from jittering. See AI operations.

Notes

• Aspect ratio — 1080×1920 (9:16) is the safe target for all three platforms.
• Order matters — resize before crop, and put fade last so it applies to the final framing.
• Silent clips underperform. Add music or narration.
• Check each platform's current duration limits before publishing and trim the segment accordingly.