How to change audio tempo — free, browser-based, no upload
- Step 1Open the tempo changer and drop your file — Drag an
MP3,WAV,FLAC,M4A,OGG,Opus, or video file onto the dropzone (the tool reads the audio track of a video). Processing is local — the FFmpeg WebAssembly core loads in the tab and the file is held in memory, never uploaded. - Step 2Set the Tempo (%) value — The single control is a Tempo (%) number input with range 25 to 400 and step 1, defaulting to 100 (no change). Enter a value above 100 to speed up, below 100 to slow down. 200 = twice as fast (half the duration); 50 = half speed (double the duration).
- Step 3Understand what the number does — The tool computes a tempo factor of
percent / 100and passes it to FFmpeg'satempofilter.atempochanges duration without touching pitch. There is no separate pitch knob here — if you want to change pitch, use the sibling pitch-shifter. - Step 4Run the conversion — Click run. FFmpeg builds the
atempochain (one stage for ordinary values, several stages for extreme ones), applies it, and re-encodes once. A 4-minute song typically finishes in a few seconds on a modern laptop; longer files scale roughly with duration. - Step 5Download the result — The output keeps your input format — an
mp3in gives anmp3out, awavin gives awavout. The file is named with a-temposuffix. Download it straight to disk; nothing was sent anywhere. - Step 6Verify duration and pitch by ear — Quick sanity check: a 100% → 125% change should cut the duration to 80% of the original (
1 / 1.25) with the pitch unchanged. If the pitch sounds shifted, you used the wrong tool —atemponever moves pitch, so a pitch change means the source itself was already off.
Tempo percentage → real result
How the single Tempo (%) control maps to duration and what stays fixed. Tempo factor = percent / 100; output duration = original / factor. Pitch is unchanged at every setting.
| Tempo (%) | Factor passed to atempo | Effect on a 4:00 file | Pitch | Typical use |
|---|---|---|---|---|
| 50 | 0.50 | Slows to 8:00 (double length) | Unchanged | Transcription, slowing a fast solo to learn it |
| 75 | 0.75 | Slows to ~5:20 | Unchanged | Gentle slowdown for practice or dictation |
| 100 | 1.00 | No change (passthrough re-encode) | Unchanged | Default — leaves tempo untouched |
| 125 | 1.25 | Speeds to ~3:12 | Unchanged | Mild tightening, comfortable faster listening |
| 150 | 1.50 | Speeds to ~2:40 | Unchanged | Noticeably faster, still fully intelligible speech |
| 200 | 2.00 | Speeds to 2:00 (half length) | Unchanged | Aggressive compression of long recordings |
| 400 | 4.00 (chained 2×2.0) | Speeds to 1:00 (quarter length) | Unchanged | Maximum supported speed-up |
Input format → output format and codec
The tempo changer preserves your source format. It re-encodes once with the codec that matches the input extension. There is no format selector in the UI for this tool.
| You drop | You get back | Encoder used | Lossy / lossless |
|---|---|---|---|
.mp3 | .mp3 | libmp3lame | Lossy (one re-encode) |
.wav | .wav | pcm_s16le | Lossless PCM |
.flac | .flac | flac | Lossless |
.m4a / .aac | .m4a | aac | Lossy (one re-encode) |
.ogg | .ogg | libvorbis | Lossy (one re-encode) |
.opus | .opus | libopus | Lossy (one re-encode) |
| Video file | Audio in matching audio format | Per audio track | Depends on track |
Tier limits that apply to the tempo changer
Tempo changer is a Pro-tier tool (it does not run on a free subscription). The file-size and duration ceilings below are the platform audio-family limits per tier.
| Tier | Max file size | Max duration / file | Files per batch | Tempo changer access |
|---|---|---|---|---|
| Free | 50 MB | 30 min | 1 | Locked — requires Pro |
| Pro | 200 MB | 120 min | 10 | Unlocked |
| Pro + Media | 100 GB | Unlimited | 100 | Unlocked |
| Developer | 100 GB | Unlimited | Unlimited | Unlocked |
Cookbook
Copy-ready recipes for the most common percentage changes. Each shows the input, the single Tempo (%) value to enter, and the resulting duration. All keep pitch fixed.
Speed a track up by a quarter (125%)
The most common mild speed-up. A 125% tempo factor of 1.25 compresses duration to 80% of the original while the pitch stays put. Useful when a track feels slightly draggy but you do not want it to sound rushed.
Input: song.mp3 duration 4:00 Set: Tempo (%) = 125 Factor: 1.25 (atempo=1.25) Output: song-tempo.mp3 duration ~3:12, same pitch
Slow a passage to half speed (50%)
A 50% tempo doubles the duration so you can hear every note. Because atempo holds pitch, a guitar solo slowed to 50% is still in the original key — far more usable for learning than a resample slowdown that drops it an octave.
Input: solo.wav duration 0:45 Set: Tempo (%) = 50 Factor: 0.50 (atempo=0.5) Output: solo-tempo.wav duration 1:30, same pitch (lossless PCM)
Double the speed of a long recording (200%)
200% halves the run time. atempo accepts up to 2.0 in a single stage, so this is one filter pass. Pitch is preserved, so a sped-up interview is still intelligible.
Input: meeting.m4a duration 1:00:00 Set: Tempo (%) = 200 Factor: 2.00 (atempo=2.0) Output: meeting-tempo.m4a duration 30:00, same voices
Push to the maximum 400% (chained stages)
FFmpeg's atempo filter only accepts 0.5–2.0 reliably per stage, so the tool chains stages for extreme factors. 400% becomes a two-stage chain. This is the ceiling exposed by the UI (max 400).
Input: lecture.mp3 duration 1:20:00 Set: Tempo (%) = 400 Factor: 4.00 -> atempo=2.0,atempo=2.0 (chained) Output: lecture-tempo.mp3 duration 20:00, same pitch
Leave tempo untouched but standardise the file (100%)
Setting 100% does no tempo change but still re-encodes through FFmpeg. This is occasionally handy to force a clean single re-encode, but note it is a lossy pass for MP3/AAC/OGG/Opus — only use it on lossless formats if you care about quality.
Input: clip.flac duration 2:10 Set: Tempo (%) = 100 Factor: 1.00 (atempo=1.0) Output: clip-tempo.flac duration 2:10, lossless re-encode
Edge cases and what actually happens
Free subscription tries to run the tool
Requires ProTempo changer is registered at minTier: pro, so a free account cannot process a file with it — the run is blocked and recorded as a tier limit, not executed. Upgrade to Pro (or higher) to unlock it. The 50 MB / 30 min free numbers are the platform audio limits, but they do not grant access to this particular tool.
You expected the pitch to change too
By designatempo never changes pitch. If you want the chipmunk/octave-down effect, that is a different operation — use pitch-shifter to move pitch in semitones, or combine pitch shift with tempo. This tool deliberately holds pitch fixed at every percentage.
You entered a value below 25 or above 400
ClampedThe Tempo (%) input enforces min 25 and max 400. Values outside that range are clamped to the bounds by the control, so you cannot accidentally request a 10x speed-up. If you genuinely need a more extreme stretch, run the tool twice (e.g. 400% then 200% = effective 800%).
Large speed-ups make fast speech sound clipped
ExpectedAt 250%+ atempo introduces audible time-domain artifacts on speech (a slight warbling on sibilants). This is inherent to the algorithm, not a bug. For transparent fast playback, 125–175% is the sweet spot; beyond ~200% expect some artifacting that grows with the factor.
Re-encoding a lossy source at 100%
Quality lossRunning an MP3/AAC/OGG/Opus through at 100% still triggers a full re-encode (the tool always decodes and re-encodes). That is one generation of lossy loss for no tempo benefit. If you only want to inspect a file without changing it, do not run it through — there is no pure passthrough.
File exceeds your tier's size or duration ceiling
RejectedA 250 MB file on Pro (cap 200 MB) or a 3-hour file on Pro (cap 120 min) is rejected before processing. Upgrade to Pro + Media for 100 GB and unlimited duration, or trim the file first with audio-trimmer and process the part you need.
Output sounds the same length as the input
Check the valueIf duration did not change, the Tempo (%) was almost certainly left at 100 (the default). Confirm the control shows your intended value before running. Remember the relationship is inverse: 200% makes the file shorter, not longer.
Very long file is slow to process
ExpectedProcessing time scales with input duration because FFmpeg decodes the whole stream, applies atempo, and re-encodes. A 1-hour file takes far longer than a 4-minute song. The tab stays responsive but the WebAssembly core is doing real CPU work locally — let it finish.
Video file dropped in
Audio onlyThe tool reads the audio track of a video and returns an audio file at the new tempo — it does not produce a re-timed video. If you need the video itself sped up with synced audio, that is a video-family operation, not this audio tool.
Result clips or distorts at the peaks
Run a limiter afteratempo can nudge peak levels slightly. If the sped-up file clips, run it through true-peak-limiter afterward to bring the ceiling down to -1 dBTP. Tempo change itself does not normalise level.
Frequently asked questions
Does changing the tempo change the pitch?
No. The tool uses FFmpeg's atempo filter, which stretches or compresses time while holding pitch constant. A 150% version sounds like the same voice talking faster, not a higher-pitched chipmunk. If you actually want pitch to move, use the pitch-shifter instead.
What is the range of the tempo control?
A single Tempo (%) number input, range 25 to 400, step 1, default 100. 100 means no change; above 100 speeds up; below 100 slows down. There are no presets or secondary controls — just that one field.
How does the percentage map to duration?
The tempo factor is percent / 100, and output duration is original / factor. So 200% (factor 2.0) halves the duration, and 50% (factor 0.5) doubles it. The relationship is inverse: a higher percent makes a shorter file.
Why does 400% work if FFmpeg's atempo only goes to 2.0?
atempo accepts roughly 0.5–2.0 per stage, so the tool chains stages for extreme factors. A factor of 4.0 becomes atempo=2.0,atempo=2.0. You just type 400 and the tool builds the chain. 400% is the maximum the control allows.
Is anything uploaded to a server?
No. FFmpeg 8.1 runs as WebAssembly in your browser tab and the file is held in memory. The audio never leaves your machine. The only server interaction is an optional usage counter (no file content) for your dashboard if you are signed in.
What output format do I get?
The same format you put in. MP3 in, MP3 out (libmp3lame); WAV in, lossless WAV out (pcm_s16le); FLAC in, FLAC out; and so on for M4A/AAC, OGG, and Opus. This tool has no format-conversion control — if you need to change format, use a converter like wav-to-mp3 before or after.
Does it set a bitrate?
No bitrate is forced for tempo changes, so MP3 output uses libmp3lame's default encode. If you need a specific bitrate, run the result through bitrate-changer afterward.
Can I run this on a free account?
No. The tempo changer is a Pro-tier tool. A free account is blocked from running it. The free audio limits (50 MB, 30 min) describe the platform floor but do not grant access to this specific tool — you need Pro or higher.
What is the largest file I can process?
Pro allows 200 MB and 120 minutes per file; Pro + Media and Developer raise that to 100 GB and unlimited duration. Files beyond your tier's ceiling are rejected before processing.
How is this different from time-stretch?
They are the same underlying atempo operation — this tool expresses it as a tempo percentage (a musician's convention), while time-stretcher expresses it as a raw factor. Pick whichever input style you prefer; the result is identical for equivalent values.
Will it work on a sped-up file I already exported?
Yes — you can stack runs. 200% on an already-200% file gives an effective 400%. But each run re-encodes lossy formats, so multiple passes on an MP3 accumulate quality loss. For big stretches, do it in one run (up to 400%) rather than chaining exports.
Can I automate tempo changes in a pipeline?
Yes. JAD's API never accepts uploads, but you can pair an @jadapps/runner on your own machine and POST to http://127.0.0.1:9789/v1/tools/tempo-changer/run with the file and { percent } option. Processing stays local on your hardware.
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.