Free tool · H.264 · Runs in your browser
Change video speed
Short answer
Choose 0.5×, 0.75×, 1.5× or 2× and the tool re-times the picture and the sound together, so pitch stays natural. On a 5.1 second excerpt, 0.5× produced 10.2 seconds of video and 2× produced 2.6. A clip with no audio track is sped up too, picture only. Nothing is uploaded.
Batch is free and unlimited - files are processed one after another on your device, with no queue or daily cap. Maximum 100 MB per file; the first run loads a ~30 MB engine once per visit.
Key takeaways
- The new length is the old length divided by the factor: our 5.067 second clip became 10.23 s at 0.5×, 6.83 s at 0.75×, 3.47 s at 1.5×, 2.63 s at 2×.
- Picture and sound do not end together to the millisecond: the video ran 0.229 s longer than the audio at 0.5× and 0.119 s longer at 2×. That is a fraction of a second at the end of the clip, not a drift through it.
- The frame rate stays 30 fps. At 0.5× that means frames are repeated rather than new ones created, and at 2× frames are dropped. The tool does no motion interpolation.
- A clip without an audio track used to fail because the filter looked for sound. The tool now retries with the picture only: the first attempt failed on our silent clip, the retry produced 2.63 s with no audio.
- Encoding time was 13 s to 29 s for the 5.1 second excerpt; slow motion takes longest because there is more video to encode.
What the tool does to your file
The picture's timestamps are scaled with setpts, and the sound is stretched or squeezed with atempo, so the two change together. FFmpeg's documentation says atempo accepts 0.5 to 100 and that tempo above 2 skips samples rather than blending them; the tool stops at 2×, so that limit is not reached.
| Setting | Value | What it means |
|---|---|---|
| Picture | setpts = 1 / factor × PTS | Frames are shown sooner or later |
| Sound | atempo = factor | Faster or slower without changing pitch |
| Choices | 0.5×, 0.75×, 1.5×, 2× | Within the range atempo handles without skipping |
| Video | H.264 (libx264), yuv420p, preset veryfast | Re-encoded at the encoder default quality |
| No audio track | Retried as picture only | Sound is not added |
Measured: four speeds on a 5 second excerpt
Measured on a 5 second 1920 × 1080 excerpt of Big Buck Bunny (CC BY 3.0, Blender Foundation) with a speech soundtrack. It is an animated film, not camera footage, so the sizes of camera video will differ. The tool's own ffmpeg arguments ran on the same single-thread WebAssembly engine your browser downloads; times are from a headless run and your device will differ.
The excerpt was 8,193,531 bytes with a 5.067 second picture and a 5.015 second soundtrack.
| Speed | Expected length | Picture length | Sound length | Output size | Frame rate | Time |
|---|---|---|---|---|---|---|
| 0.5× | 10.00 s | 10.233 s | 10.004 s | 2,013,637 bytes (2.01 MB) | 30 fps | 28.8 s |
| 0.75× | 6.67 s | 6.833 s | 6.661 s | 1,717,488 bytes (1.72 MB) | 30 fps | 22.2 s |
| 1.5× | 3.33 s | 3.467 s | 3.340 s | 1,306,513 bytes (1.31 MB) | 30 fps | 14.9 s |
| 2× | 2.50 s | 2.633 s | 2.514 s | 1,162,836 bytes (1.16 MB) | 30 fps | 12.8 s |
The Change video speed tool on a 5 second 1080p excerpt.
The expected length is the 5 second excerpt divided by the factor. The picture is a little longer than that because the encoder rounds up to whole frames at the end; the sound ends nearer the expected time.
Silent clips
A video with no audio track used to make the whole job fail, because the filter graph asked for a sound stream that was not there. The tool now catches that and runs again for the picture only. Our measurement: the first command failed on the silent excerpt, and the second produced a 2.633 second file with no audio in 11.9 s.
Slow motion and fast motion at a fixed frame rate
The file keeps its original frame rate. At 2× the tool drops every other frame; at 0.5× it shows each frame twice. Smooth slow motion from ordinary footage needs frames that were never recorded, which ffmpeg can create with a separate interpolation filter. This tool does not use it, so very slow speeds look steppy.
For a smoother result, film at a higher frame rate, or use a 0.75× setting, which repeats frames less often.
Free tools:Trim by timestampCompress a videoMute a videoWhat is frame rate?
Planning a length
- Divide the length you have by the speed: a 60 second clip at 1.5× is 40 seconds; at 0.5× it is 120.
- Want a given length? Divide the length you have by the length you want: 90 seconds down to 30 is 3×, which is beyond one run of this tool, so run 2× and then 1.5× (2 × 1.5 = 3), remembering that each run re-encodes.
- Speed up before you compress, so the encoder works on the shorter file.
Questions people ask
How do I speed up a video?
Choose 1.5× or 2×. The picture and the sound both play faster, and the pitch is kept.
How do I slow a video down?
Choose 0.5× or 0.75×. On our excerpt 0.5× doubled the length to 10.2 seconds. The frame rate stays 30 fps, so frames are repeated, not interpolated.
Does the sound change pitch?
No. The sound is time-stretched with atempo, which changes tempo and not pitch.
Does it work on a video with no sound?
Yes. The tool retries with the picture only: our silent excerpt came out 2.63 seconds long with no audio.
Why is the picture slightly longer than the sound?
The encoder rounds up to whole frames. On our clip the picture ran 0.229 seconds longer at 0.5× and 0.119 seconds longer at 2×.
Can I go faster than 2×?
Not in this tool. Run it twice for 4×, remembering each run re-encodes.
Is anything uploaded?
No. Files up to 100 MB are processed in your browser.
Related
Sources and how we checked
- FFmpeg Filters Documentation (read 2026-10-06)
- FFmpeg wiki: Encode/H.264 (CRF and presets) (read 2026-10-06)
Reviewed October 9, 2026. The measurements ran the tool’s own ffmpeg arguments on the same single-thread WebAssembly engine your browser downloads. The script, the results and the saved documentation are in data/seo/hub/evidence/2026-10-09/tools.