How to speed up or slow down audio without pitch shift
- Step 1Drop in the audio you want to re-speed — Add a lecture, podcast, audiobook chapter, song, or clip (mp3, m4a, wav, flac, ogg, opus, aiff). It loads into FFmpeg WASM locally — no upload.
- Step 2Decide the direction — Speeding up (save time) → factor above 1. Slowing down (study) → factor below 1. The same Factor input does both.
- Step 3Pick a value that stays clear — Speed up: 1.25–1.5 keeps speech crisp. Slow down: 0.6–0.85 keeps music and speech transparent. The further from 1, the more artefacts.
- Step 4Enter the factor — Type it in the Factor input (0.25–4, step 0.05). Output duration = original ÷ factor: 1.5 makes it ~33% shorter; 0.75 makes it ~33% longer.
- Step 5Process locally — FFmpeg applies
atempo=<factor>and re-encodes in your input format. Pitch is preserved in both directions. - Step 6Download the re-sped file — Save the
-stretchedfile. Now you have a faster lecture to commute with, or a slowed solo to practise against — as a file, not just a setting.
Both directions at a glance
Speed-up factors (above 1) shorten audio; slow-down factors (below 1) lengthen it. Pitch is preserved in every row.
| Goal | Direction | Suggested factor | Effect on length |
|---|---|---|---|
| Skim a podcast | Speed up | 1.50 | ~33% shorter |
| Save time on a lecture | Speed up | 1.25 | 20% shorter |
| Audiobook at a brisk pace | Speed up | 1.30 | ~23% shorter |
| Comfortable clarity | Unchanged | 1.00 | Same length |
| Study a music passage | Slow down | 0.70 | ~43% longer |
| Transcribe / dictation | Slow down | 0.80 | 25% longer |
Factor → speed → duration
Full factor reference covering both directions. Output duration is always the original divided by the factor.
| Factor | Playback speed | Output duration | atempo stages |
|---|---|---|---|
| 0.25 | Quarter speed (slowest) | 4× longer | Two (0.5 × 0.5) |
| 0.5 | Half speed | 2× longer | One |
| 0.75 | 75% speed | 1.33× longer | One |
| 1.0 | Unchanged (default) | Same length | One (no-op) |
| 1.25 | 125% speed | 0.8× (20% shorter) | One |
| 1.5 | 150% speed | 0.667× (33% shorter) | One |
| 2.0 | Double speed | Half length | One |
| 4.0 | Quadruple speed (fastest) | Quarter length | One |
Tier limits for the time-stretcher
Audio-family limits read from the code (lib/tier-limits.ts). The per-file DURATION cap is separate from the file-size cap — a 35-minute voice memo well under 50 MB still exceeds the Free 30-minute limit. The time-stretcher is a Pro-tier tool (minTier: pro).
| Tier | Max file size | Max duration per file | Files per batch |
|---|---|---|---|
| Free | 50 MB | 30 min | 1 (time-stretcher itself starts at Pro) |
| Pro | 200 MB | 120 min | 10 |
| Pro-media | 100 GB (streaming) | Unlimited | 100 |
| Developer | 100 GB | Unlimited | Unlimited |
Cookbook
Jobs in both directions — faster for review, slower for study. All keep pitch constant; factors round to the 0.05 step.
Speed up a 60-minute lecture to 45 minutes
A recorded lecture is reviewed faster at 1.33×, with the lecturer's voice still natural (no chipmunk).
Input : lecture.mp3 (60 min) Factor : 1.33 (133% speed) Output : 60 / 1.33 = 45.1 min Same pitch; ~15 minutes saved.
Slow a song to 0.7× to learn the melody
A fast vocal line is slowed to 70% to study phrasing — pitch unchanged, so you can sing/play along in the same key.
Input : track.flac Factor : 0.70 Output : track-stretched.flac (1.43x longer, lossless) Key unchanged -- practise in the original tuning.
Audiobook chapter at 1.25× for the commute
A chapter is sped up 25% to fit a commute, narration still clear and in the narrator's natural voice.
Input : chapter.m4a (40 min) Factor : 1.25 Output : 40 / 1.25 = 32 min (encoder: aac)
Slow a podcast guest's fast section for transcription
One guest talks fast. Trim that section, slow it to 0.8×, and type along — pitch preserved keeps the voice recognisable.
Step 1: audio-trimmer -> fast-section.mp3 (4 min) Step 2: time-stretcher factor 0.80 -> 5 min Type against the slowed, natural-pitch copy.
Round-trip a clip both ways to compare
Make a faster and a slower copy of the same clip to pick the most usable pace, without ever re-pitching.
Source : clip.wav (30 s) Fast copy : factor 1.40 -> 21.4 s (skim) Slow copy : factor 0.80 -> 37.5 s (study) Both keep the original pitch; choose per use.
Edge cases and what actually happens
Speeding up made it sound like a chipmunk
Wrong setting / toolTime-stretch preserves pitch, so it shouldn't chipmunk. If your audio is detuning, you're using a resample/speed tool, not this one — or you meant to change pitch, in which case use pitch-shifter.
Slowed audio dropped in pitch
Wrong toolPitch dropping on slow-down is the resample behaviour, not atempo. This tool holds pitch constant; if you saw a key change, you weren't using the time-stretcher's slow-down. Re-run here with a factor below 1.
Sped-up speech is hard to follow above 1.5×
Algorithm limitatempo stays clear to about 1.5× for speech; beyond that the time-stretch artefacts and the sheer pace reduce intelligibility. For comprehension, keep speed-ups at 1.25×–1.5×.
Slow-down below 0.5× sounds smeary
Algorithm limitBelow 0.5× the tool chains atempo stages and the stretch becomes audibly watery. 0.5–0.85× is the transparent zone for study; deeper is for effects.
Output longer/shorter than you expected
By designFactor is a speed multiplier: above 1 shortens, below 1 lengthens, and output = original ÷ factor. If the length went the 'wrong' way, you picked the wrong side of 1.
Long lecture exceeds the duration cap
RejectedA 130-minute lecture exceeds Pro's 120-minute cap (and Free's 30). Use Pro-media for unlimited duration, or split it first with audio-splitter and speed up each part.
MP3 podcast/lecture re-encoded on speed-up
Re-encode generationStretching re-encodes the audio, adding one lossy generation to an MP3. For listening that's inaudible. It's only worth caring about for archival masters.
You only need a player setting, not a file
Consider a playerIf you just want to listen faster once, a podcast app's speed control does that without re-encoding. Use this tool when you need a saved file to loop, share, or import elsewhere.
Frequently asked questions
Can one tool both speed up and slow down audio?
Yes. The single Factor control does both: above 1 speeds up and shortens, below 1 slows down and lengthens. Pitch is preserved in both directions, so neither chipmunks nor detunes.
How do I speed up a lecture without the chipmunk effect?
Enter a factor above 1 (1.25–1.5 works well). FFmpeg's atempo speeds up the tempo while keeping pitch constant, so the lecturer sounds natural — just faster. Output length = original ÷ factor.
How do I slow down music without detuning it?
Enter a factor below 1 (0.6–0.85 for study). atempo slows the tempo but holds pitch, so the song stays in its original key while you learn the part.
What's the best factor for podcasts and audiobooks?
1.25×–1.5× keeps speech clear while saving meaningful time. Above ~1.5× the words get harder to follow because of pace and time-stretch artefacts.
What's the best factor for music practice?
0.6×–0.85× is the sweet spot — slow enough to study, transparent enough to sound right. Below 0.5× the audio gets smeary as atempo chains stages.
Does this give me a file or just a setting?
A real downloadable file in your input format. Unlike a player's speed slider, you can keep it, loop it, share it, or import it into a DAW or transcription tool.
Will the format change?
No. There's no format dropdown; the output keeps the input extension and is re-encoded with that format's encoder (mp3→libmp3lame, m4a→aac, wav→pcm_s16le, etc.).
Is my course or podcast audio uploaded?
No. Processing runs in your browser via FFmpeg WebAssembly. Course recordings and private podcasts stay on your machine; only an anonymous job counter is logged when signed in.
How much time do I save speeding up?
Time saved = original × (1 − 1 ÷ factor). At 1.25× you save 20%; at 1.5× you save ~33%; at 2× you save 50% (though speech gets hard to follow that fast).
How long a file can I process?
This is a Pro feature: Pro allows 120 minutes per file, Pro-media and Developer remove the duration cap (up to 100 GB). Split very long lectures with audio-splitter if needed.
Should I use pitch-shifter or tempo-changer instead?
Use pitch-shifter to change key without changing length. Use tempo-changer to set tempo as a percentage instead of a multiplier. Time-stretcher changes length (both ways) while keeping pitch.
Why does going far above or below 1× sound off?
atempo is a time-domain algorithm; it's transparent near 1× and adds smearing/ringing as you push further. Keep speed-ups under ~1.5× and slow-downs above ~0.5× for the cleanest results.
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.