How to speed up video 2× for tutorials and walkthroughs
- Step 1Drop the screen recording in — Accepts MP4, MOV, MKV, WebM, AVI, M4V, TS — covering OBS, ScreenFlow, Camtasia, and QuickTime exports. The file is read locally; nothing uploads. Speed change is a Pro feature (£7/month). The file card shows duration so you can plan the target runtime.
- Step 2Set Speed to your target pace — For tutorials,
2halves the runtime and still keeps speech clear;1.5is gentler and very safe for narration;1.25barely registers as faster. Type the value in the Speed (× · 0.25 to 4) field. The default is1. - Step 3Estimate the new runtime — Output duration ≈ input ÷ speed. A 40:00 recording at 2× → 20:00; at 1.5× → ~26:40; at 1.25× → 32:00. Pick the speed that lands near the runtime your platform or audience prefers.
- Step 4Run and let atempo handle the audio — Click Run Speed. FFmpeg builds
[0:v]setpts=...[v];[0:a]atempo=...[a]— a single atempo pass covers 1.25×–2×, so narration gets the cleanest possible time-stretch in this range. Output re-encodes to H.264 CRF 20 + AAC 192k. - Step 5Download and spot-check intelligibility — Play a dense section — fast typing while you explain a concept. At 2×, words can run together; if so, re-run at 1.75× or 1.5×. The result is a normal MP4 download (no streaming-to-disk for this tool).
- Step 6Trim dead air first for an even tighter cut — Speeding up shrinks pauses too, but they're still there. For a tighter result, cut long silent gaps with the lossless trimmer or split out the section you need with the video splitter before applying speed.
Recommended speed by tutorial length and audio density
Guidance for instructional narration. atempo stays a single pass up to 2× (cleanest). Beyond 2× it chains and speech gets harder to follow.
| Source runtime | Speed 1.25× | Speed 1.5× | Speed 2× | Best for |
|---|---|---|---|---|
| 10:00 | 8:00 | 6:40 | 5:00 | Short how-to clips |
| 20:00 | 16:00 | 13:20 | 10:00 | Feature walkthroughs |
| 40:00 | 32:00 | 26:40 | 20:00 | Full lessons / modules |
| 1:00:00 | 48:00 | 40:00 | 30:00 | Long-form courses |
| Audio density | Dense narration — safe | Most tutorials — safe | Sparse / slow speech — safe | Match speed to talk speed |
Speed value vs. filter chain (tutorial range)
The exact FFmpeg filters the tool builds for common tutorial speeds. A single atempo pass keeps narration cleanest.
| Speed | setpts (video) | atempo (audio) | atempo passes |
|---|---|---|---|
1.25 | 0.8000*PTS | atempo=1.2500 | 1 (clean) |
1.5 | 0.6667*PTS | atempo=1.5000 | 1 (clean) |
1.75 | 0.5714*PTS | atempo=1.7500 | 1 (clean) |
2 | 0.5000*PTS | atempo=2.0000 | 1 (clean, at the limit) |
3 | 0.3333*PTS | atempo=2.0,atempo=1.5 | 2 (chained — words run together) |
Cookbook
Practical recipes for tightening screencasts, course modules, and software demos while keeping the narrator clear.
40-minute lesson halved to 20 minutes
The classic course-trimming pass. 2× keeps narration clear because it's a single atempo stage, and viewers are far more likely to finish a 20-minute lesson than a 40-minute one.
Speed field: 2 FFmpeg filter graph: [0:v]setpts=0.5000*PTS[v];[0:a]atempo=2.0000[a] 40:00 input -> 20:00 output (H.264 MP4, AAC 192k) Narration pitch unchanged, just faster delivery
Gentle 1.5× when 2× clips the words
For dense narration where 2× makes words blur, 1.5× still cuts a third off the runtime with very natural-sounding speech.
Speed field: 1.5 [0:v]setpts=0.6667*PTS[v];[0:a]atempo=1.5000[a] 30:00 input -> 20:00 output Feels like a presenter who just talks a bit faster
Trim dead air, then speed up
Speed shrinks pauses but doesn't remove them. Cut the long silent gaps first, then apply 2× for the tightest possible tutorial.
Step 1: lossless-trimmer -> remove 3 min of setup silence
25:00 -> 22:00
Step 2: speed-controller, Speed 2
22:00 -> 11:00 final
Two tools, one tight walkthrough.Speed only the boring middle of a demo
Keep intro and conclusion at normal pace, speed only the slow build step. Split, speed the middle, rejoin.
Step 1: video-splitter -> intro | build | wrap-up Step 2: speed-controller on 'build' only, Speed 2.5 Step 3: video-merger -> stitch back together Result: natural intro/outro, fast-forwarded grind in the middle.
Why 3× starts to hurt narration
Above 2×, atempo chains, and the time-compression makes consecutive words overlap. For instruction, 2× is the practical ceiling for clear speech.
Speed field: 3 [0:a]atempo=2.0,atempo=1.5[a] (2 chained passes) Result: runtime is great, but technical terms can blur. Drop to 2x (or 2.25x) if comprehension matters.
Edge cases and what actually happens
2× makes dense technical narration blur
Expected2× is the top of the single-atempo-pass range and is clean for most speech, but rapid-fire narration with jargon can run together at that rate. Drop to 1.75× or 1.5× — both are still single-pass and noticeably tighter than real time. There is no per-word smoothing; the fix is choosing a gentler factor.
Screen recording has no audio (silent capture)
FailsSome screen recordings are captured without a mic. The filter graph always maps an audio stream through atempo, so a truly silent file fails with a stream-mapping error. Record with audio enabled, or add a silent track in another tool before speeding up.
Code/UI text still readable after speed-up?
Preservedsetpts changes only the timing, never the resolution, so on-screen code, menus, and cursor stay exactly as sharp as the source. The only caveat is the CRF 20 re-encode, which is visually high quality; fine text in a 1080p+ screencast remains legible.
Want faster than 4× for a very long capture
Two passesA 2-hour unattended capture can't go straight to, say, 8× — the per-run ceiling is 4×. Run at 4×, then feed the output back at 2× for an effective 8×. Each pass re-encodes, so chain only as many times as you actually need.
Variable-frame-rate screen capture (OBS / browser tab)
SupportedScreen recorders often produce variable frame rates. setpts re-times by timestamp and the export is re-encoded H.264, so VFR captures come out synced and play consistently. If you saw audio drift in other editors with the same file, this combined graph keeps it locked.
Output is MP4 even from a MOV/MKV screencast
By designThe export is always H.264 MP4 + AAC regardless of input container. That's ideal for re-uploading tutorials to YouTube/LMS. If your pipeline specifically needs MKV or WebM, run the MP4 through the transcoder afterwards.
Free tier — run button disabled
Pro requiredSpeeding up video needs a Pro subscription (£7/month). On free, the control is gated behind an upgrade prompt. File-size limits: Free 1 GB / 1 file, Pro 10 GB / 5 files, Pro + Media 100 GB / 50 files — no duration cap, only size and batch-count.
Batch of module recordings at one speed
Supported (batch)Drop several lesson files together and they queue as a batch (up to 5 on Pro, 50 on Pro + Media). The same Speed applies to all of them — convenient when every module of a course needs the identical 1.5× or 2× treatment.
Leaving Speed at 1 to 'just clean up' the file
Re-encodes anywaySpeed 1 still runs the filters and re-encodes; it does not stream-copy. If you only want to shrink or web-optimise a tutorial without changing pace, use the web optimizer, which sets faststart and a sensible H.264 encode for streaming.
Long 1080p+ screencast on a modest laptop
May be slowRe-encoding an hour of high-resolution screen capture in the browser is CPU-heavy. It will complete within your tier limit, but expect a wait. Trimming or splitting to just the part you're publishing first keeps the speed pass quick.
Frequently asked questions
Will my narrator sound like a chipmunk at 2×?
No. The audio goes through FFmpeg's atempo time-stretch, which speeds up delivery without raising pitch. A 2× tutorial sounds like a presenter talking faster, not a higher-pitched voice — exactly what you want for instructional content where clarity matters.
What speed is best for a tutorial?
2× is the sweet spot for clear speech and is still a single atempo pass (cleanest). If narration is dense or fast, drop to 1.5× or 1.75×. Above 2× atempo starts chaining and words can run together, so 2× is the practical ceiling for comprehension.
How much shorter will my video get?
Roughly input duration divided by the speed. A 40-minute lesson is 20 minutes at 2×, about 27 minutes at 1.5×, and 32 minutes at 1.25×. Pick the speed that lands near the runtime your audience or platform expects.
Does speeding up blur on-screen text or code?
No. The speed filter (setpts) changes only the timing, never the resolution — your menus, code, and cursor stay as sharp as the original. The export re-encodes at CRF 20, which keeps fine text in a 1080p+ screencast legible.
Can I speed up only the slow part of a demo?
Not within this tool — speed applies to the whole clip. To keep the intro/outro normal and fast-forward the middle, split the recording (try the video splitter), speed only that segment, then merge the pieces back together.
Is my course material uploaded anywhere?
No. Everything runs in your browser via WebAssembly FFmpeg. Pre-launch course videos and client demos are read locally and never sent to a server — only an anonymous 'file processed' counter is recorded for dashboard stats.
What format do I get back?
A standard H.264 MP4 (libx264 CRF 20, AAC 192 kbps) no matter what you put in. That re-uploads cleanly to YouTube, Vimeo, Teachable, Thinkific, or any LMS. Need a different container? Run it through the transcoder.
My screen recording is silent — why won't it process?
The tool always routes an audio stream through atempo. A capture made without a microphone has no audio track for it to process, so the job fails. Record with audio on, or add a silent track first. Tutorials with narration are unaffected.
Can I batch a whole course at once?
Yes. Drop multiple module files together and they queue as a batch — up to 5 files on Pro or 50 on Pro + Media. The same speed is applied to every file, which is handy when each module needs the same 1.5× or 2× treatment.
Should I trim before or after speeding up?
Trim first if you can. Speeding up shrinks pauses proportionally but doesn't delete dead air. Removing long silent gaps with the lossless trimmer before the speed pass gives the tightest final tutorial.
How big a file can I process?
Free allows up to 1 GB / one file, Pro up to 10 GB / five files per batch, Pro + Media up to 100 GB / fifty files. There's no minutes limit — only file size and batch count. Speed control itself requires the Pro tier (£7/month).
Does leaving speed at 1 do anything useful?
It re-encodes the file to H.264/AAC MP4 but doesn't re-time anything. If you only want to optimise a tutorial for streaming (faststart, sensible bitrate) without changing pace, use the web optimizer instead.
Privacy first
Every JAD Video tool runs entirely in your browser via WebCodecs and FFmpeg (WebAssembly). Your video files never leave your device — verified by zero outbound network requests during processing.