Make a Timelapse From Video

Pick a video, a speed-up factor, and an output fps — frames are dropped and sped up in your browser. Audio is removed, nothing is uploaded.

Try:
Timelapse video

About this tool

Turn long, slow footage — a sunset, a build, a drive, a whiteboard session — into a short, fast timelapse, without leaving your browser. Pick a video, choose how many times faster it should play (say 10×), set an output frame rate, and the clip is compressed in time locally. The file never leaves your device.

How it works

The tool runs two ffmpeg filters compiled to WebAssembly. First setpts divides each frame's timestamp by your speed factor (setpts=PTS/10), so a 60-second clip is squeezed into 6 seconds. That squeeze crams every original frame into a tiny window, so a second fps filter then drops the surplus frames down to your chosen output rate — that frame-drop is exactly what makes a timelapse look smooth and stay small instead of a stuttering all-frames blur.

The sped-up video is re-encoded to H.264 at -crf 20 (visually near-transparent quality), forced to yuv420p with +faststart so it plays and scrubs everywhere. Audio is always removed — a 10×-fast soundtrack is just noise, and dropping it keeps the output small. An .mp4, .mov, .m4v, or .mkv keeps its container; other inputs (.webm, .ogv, .avi, …) come out as MP4.

Example

Upload a 2-minute clip of clouds rolling past, set the speed to 20× and the output to 30 fps, and you get a 6-second timelapse of the same scene — roughly 120s ÷ 20 = 6s — silent, small, and ready to share.

Notes

FAQ

How do I decide what speed to use?

Work backwards from how long you want the result. Output length is roughly the input length divided by the speed, so a 5-minute clip at 10× is about 30 seconds, and at 20× about 15 seconds. For a very long recording (an hour or more) pick a high factor like 60× or beyond; the default is 10×. Any value from to 300× is accepted and clamped into that range.

Why is there no sound on the result?

A timelapse plays many times faster than real time, so the original audio would be an unintelligible high-speed rush. The tool removes the audio track on purpose — this also keeps the output file small. If you actually want to speed up a clip and keep its audio pitch-corrected, that's a speed change rather than a timelapse.

What does the output fps control do?

After the clip is sped up, it's re-sampled to the frame rate you pick — 30 (web/general), 24 (a cinematic film look), 25 (PAL), or up to 60 for extra-smooth motion. Lowering it drops more frames (smaller file); the default is 30. It does not add new frames — raising it just repeats existing ones.

Will a timelapse look smooth or choppy?

Because frames are dropped rather than blended, very high speed factors on already-jerky footage can look stepped. For the smoothest result, shoot or start from steady, high-frame-rate footage and keep the output fps at 30 or higher. This tool does not do motion interpolation, which is what would smooth out extreme speed-ups.

Does the file get uploaded anywhere?

No. The whole re-encode runs on your own CPU via ffmpeg compiled to WebAssembly, so the video stays on your device. That also means encoding time grows with the clip's duration and resolution — a long 4K recording can take a few minutes, while a short phone clip finishes in seconds.

Developer & Automation Access

Run it from the terminal

Same engine as this page, headless — via the gizza CLI:

gizza tool video-timelapse 'url=https://example.com/input' 'speed=10' 'fps=30'

New to the CLI? Get gizza →

Open it by URL

Pre-fill and auto-run this tool with query parameters — the names match the API/CLI:

https://gizza.ai/tools/video-timelapse/?url=https://example.com/input&speed=10&fps=30

Machine-readable descriptor: tool.json — title + parameters JSON Schema for agents.