How to speed up or slow down audio online — free no upload
- Step 1Drop the recording — Drag your
MP3,WAV,FLAC,M4A,OGG,Opus, or video file onto the dropzone. The FFmpeg WebAssembly core loads in the tab; the file stays in browser memory and is never uploaded. - Step 2Decide: speed up or slow down — There is one Tempo (%) control. Above 100 speeds the recording up (shorter file); below 100 slows it down (longer file). 100 is no change. The same field handles both directions.
- Step 3Pick a value — Common choices: 60–75% to slow speech for transcription; 125–150% to comfortably skim-listen; 200% to halve a long meeting; 50% to learn a fast musical passage. The control accepts 25 to 400, step 1.
- Step 4Run it — Click run. FFmpeg applies
atempo(factor =percent / 100), chaining stages automatically for extreme values, and re-encodes once into your source format. - Step 5Download the new file — Save the result, which carries a
-temposuffix. It is a standalone file at the new speed — load it into a transcription tool, send it to a colleague, or keep it as your reference copy. - Step 6Confirm the speed by checking duration — A 200% file should be half the original length; a 50% file double. If the duration looks wrong, re-open and check the Tempo (%) value — the most common mistake is leaving it at the default 100.
Speed-up and slow-down quick reference
What each Tempo (%) value does to a recording's length and feel. Pitch is held constant for every setting. Factor = percent / 100; new duration = original / factor.
| Goal | Tempo (%) | 10:00 file becomes | Best for |
|---|---|---|---|
| Deep slowdown | 50 | 20:00 | Transcription, learning a fast solo note-by-note |
| Moderate slowdown | 70 | ~14:17 | Dictation, slowing a fast talker |
| Slight slowdown | 90 | ~11:07 | Taking notes while you listen |
| No change | 100 | 10:00 | Passthrough (re-encode only) |
| Comfortable speed-up | 130 | ~7:42 | Skim-listening that still sounds natural |
| Strong speed-up | 175 | ~5:43 | Fast review of familiar material |
| Half the time | 200 | 5:00 | Halving a long meeting or interview |
Listening intelligibility by speed range
Practical guidance on where atempo stays transparent versus where artifacts appear. atempo is time-domain, so very fast speech develops a slight warble.
| Tempo (%) range | Speech intelligibility | Artifact level | Notes |
|---|---|---|---|
| 50–90 | Very high | Minimal | Slowdowns are the cleanest; great for transcription |
| 100–150 | Very high | Minimal | The transparent speed-up sweet spot |
| 150–200 | High | Slight | Still fully usable; minor warble on sibilants |
| 200–300 | Moderate | Noticeable | Fine for familiar content, less so for first listen |
| 300–400 | Lower | Strong | Maximum speed; expect audible time-stretch artifacts |
Tier limits for speeding/slowing files
The tempo changer is a Pro-tier tool. Size and duration ceilings are the platform audio-family limits.
| Tier | Max file size | Max duration / file | Access |
|---|---|---|---|
| Free | 50 MB | 30 min | Locked (needs Pro) |
| Pro | 200 MB | 120 min | Unlocked |
| Pro + Media | 100 GB | Unlimited | Unlocked |
| Developer | 100 GB | Unlimited | Unlocked |
Cookbook
Practical speed-up and slow-down recipes for spoken-word and music files. Each lists the exact Tempo (%) to enter and the resulting length.
Slow a fast speaker to 70% for transcription
Court recordings, fast-talking interviewees, and rapid lectures are easier to transcribe slowed down. 70% stretches a 10-minute clip to roughly 14:17 with the voice still natural — no octave drop to make words harder to parse.
Input: interview.mp3 duration 10:00 Set: Tempo (%) = 70 Factor: 0.70 Output: interview-tempo.mp3 duration ~14:17, same voice
Speed an audiobook to 150% and keep it
If you like listening at 1.5x but want a permanent faster copy (for a player without speed control, or to share), bake it in. 150% turns an 8-hour book into roughly 5:20 of audio at the original pitch.
Input: chapter.m4a duration 1:00:00 Set: Tempo (%) = 150 Factor: 1.50 Output: chapter-tempo.m4a duration 40:00, same narrator pitch
Halve a long meeting recording (200%)
A two-hour all-hands you only need to skim. 200% halves it to one hour. atempo handles 2.0 in a single stage, and pitch stays put so the speakers are still recognisable.
Input: allhands.wav duration 2:00:00 Set: Tempo (%) = 200 Factor: 2.00 Output: allhands-tempo.wav duration 1:00:00 (lossless PCM)
Slow a guitar lick to 50% to learn it
Musicians slow recordings to half speed to catch fast passages. Because atempo keeps pitch, the lick stays in the original key — you can play along without re-tuning.
Input: lick.wav duration 0:30 Set: Tempo (%) = 50 Factor: 0.50 Output: lick-tempo.wav duration 1:00, same key
Make a 1.3x skim copy that still sounds natural
130% is the comfortable speed-up most people do not consciously notice. Good default for review copies of podcasts or training audio where you want speed without artifacts.
Input: training.mp3 duration 45:00 Set: Tempo (%) = 130 Factor: 1.30 Output: training-tempo.mp3 duration ~34:37
Edge cases and what actually happens
Free account cannot export
Requires ProSpeeding or slowing a file with this tool needs a Pro subscription — the tool is registered at minTier: pro. A free account's run is blocked and recorded as a tier limit rather than producing a file. Upgrade to Pro to use it.
Slowed-down speech sounds the same pitch — is that right?
By designYes. That is the whole point of atempo: it slows time without dropping pitch, which is exactly what you want for transcription. If you wanted the lower-pitched, tape-slowdown sound, that is a pitch operation — see pitch-shifter.
Fast speed-up develops a warble
ExpectedAbove roughly 200% atempo's time-domain stretching becomes audible as a slight warble on sibilants and transients. It is the algorithm, not your file. Keep speed-ups under ~200% for transparent results; 130–175% is ideal for skim-listening.
You typed 500% hoping for an extreme skim
Clamped to 400The control caps at 400. For more, run the output through the tool again (e.g. 400% then 150% ≈ 600%), but expect heavy artifacts and accumulating lossy re-encode loss on MP3/AAC.
Recording is over your tier's duration cap
RejectedA 3-hour file on Pro (120 min cap) is rejected. Either trim it first with audio-trimmer, split it with audio-splitter and process the parts, or upgrade to Pro + Media for unlimited duration.
Output is the same length as the input
Check the valueTempo (%) was left at 100 (no change). Set it to your intended speed before running. Remember: 200% makes the file shorter, 50% makes it longer — the percentage is inverse to duration.
MP3 sounds slightly degraded after slowing
One re-encodeAny tempo change re-encodes the file once. For a lossy MP3 that is one generation of loss — usually inaudible at a single pass. If quality is critical (e.g. archival transcription source), work from a WAV or FLAC original so the slow-down stays lossless.
Processing a long file takes a while
ExpectedA two-hour recording decodes, re-times, and re-encodes entirely in your browser — that is real CPU work and scales with duration. The tab stays usable; let the WebAssembly core finish rather than reloading.
You dropped in a video to slow the audio
Audio onlyThe tool extracts and re-times the audio track and returns an audio file. It does not output a slowed video. For synced slow-motion video you need a video tool, not this one.
Sped-up file peaks slightly hotter
Limit if neededSpeed changes can nudge peaks. If the result clips, run true-peak-limiter to set a -1 dBTP ceiling afterward.
Frequently asked questions
Can I both speed up and slow down with this tool?
Yes, with the same single Tempo (%) control. Values above 100 speed up; values below 100 slow down. 100 is no change. The field accepts anything from 25 to 400.
Does slowing the audio lower the pitch?
No. atempo slows time while keeping pitch constant, so a slowed recording sounds like the same person speaking slower, not a deep tape-slowdown. This is ideal for transcription, where a pitch drop would make words harder to hear.
How do I turn a percentage into a duration?
New duration = original / (percent / 100). So 200% halves the length and 50% doubles it. The percentage is inverse to duration: higher percent, shorter file.
What speed is best for skim-listening without artifacts?
Roughly 130–175% stays transparent for most speech. Above 200% you start to hear a slight warble from atempo's time-stretching. For first-time listening of unfamiliar material, stay near 150%.
Is the recording uploaded?
No. FFmpeg runs as WebAssembly in your browser and the file stays in memory locally. Nothing is sent to a server, which matters for confidential meetings and interviews. Only an anonymous usage counter is logged if you are signed in.
What format will the exported file be?
The same as your input. MP3 in gives MP3 out, WAV in gives lossless WAV out, and likewise for FLAC, M4A, OGG, and Opus. To change format, convert before or after with a tool like mp3-to-wav.
Do I need a paid plan?
Yes — the tempo changer is a Pro-tier tool. Free accounts are blocked from running it. Pro and above unlock it, with size/duration limits rising by tier.
How long a recording can I process?
Pro allows up to 120 minutes and 200 MB per file. Pro + Media and Developer remove the duration limit and allow up to 100 GB. Files past your tier's cap are rejected before processing.
Can I slow down a song to transcribe lyrics?
Yes — 60–75% slows vocals enough to catch words while keeping them in the original key (atempo preserves pitch). For tricky sections, 50% gives the most headroom. Use a lossless source (WAV/FLAC) for the cleanest result.
Will repeated speed changes degrade the file?
Each run re-encodes once. For lossless WAV/FLAC that is harmless; for MP3/AAC each pass adds a little loss. Do large speed changes in one run (the control supports up to 400%) rather than stacking exports.
What is the difference between this and a media player's speed button?
A player only changes playback on that device temporarily. This tool bakes the speed into a new file you can download, share, and reuse anywhere. Same idea, but you get a permanent artifact.
Can I batch many recordings at once?
Pro allows up to 10 files per batch, Pro + Media up to 100, Developer unlimited. For automated runs, pair an @jadapps/runner and POST files plus { percent } to http://127.0.0.1:9789/v1/tools/tempo-changer/run — processing stays on your own machine.
Privacy first
Every JAD Audio tool runs entirely in your browser via FFmpeg (WebAssembly) and RNNoise. Your audio files never leave your device — verified by zero outbound network requests during processing.