Endre hastighet
Raskere eller saktere
Om dette verktøyet
Speed up or slow down an audio file without making it sound chipmunky or boomy. Privvert uses FFmpeg's atempo filter, which changes the tempo while keeping the original pitch - so a podcast at 1.5× still sounds like the host, just faster.
It's the right tool for cramming a long lecture into a commute, slowing down a fast tutorial to take notes, practicing a piece of music at a comfortable tempo, or learning a foreign language by listening to native speech at 0.75×. Everything runs offline in your browser.
Under the hood the atempo filter divides the signal into short overlapping windows, stretches or compresses each one in the time domain and crossfades them back together. The pitch is left alone because the frequency content of each window is preserved - only the spacing between windows changes. For ratios outside the 0.5×-2× range, atempo is automatically chained (e.g. 0.5 × 0.5 = 0.25×) so artifacts stay minimal.
Funksjoner
- 0.5× to 2× speed range with one click
- Pitch-preserving tempo change (no chipmunk effect)
- Optional pitch-shifting mode for creative use
- Reads MP3, WAV, M4A, AAC, OGG, FLAC, Opus
- Outputs MP3 at 192 kbps
- Local processing - files never uploaded
- Free and unlimited
Slik bruker du det
- Drop in an audio file.
- Pick a speed multiplier (e.g. 1.25× for podcasts, 0.85× for transcription).
- Click Apply.
- Download the time-stretched MP3.
Alt skjer inne i nettleseren din med JavaScript og WebAssembly. Filene dine lastes aldri opp, lagres aldri og ses aldri av oss.
Ofte stilte spørsmål
We use the atempo filter, which uses time-stretching algorithms (overlap-add) to change tempo while preserving pitch. Old-school speed-up just played the file faster, which raised the pitch - atempo doesn't.
atempo is reliable from 0.5× to 2× per pass. The tool automatically chains passes if you request a more extreme ratio (e.g. 0.25×).
Yes - half the length means roughly half the file size at the same bitrate.
Switch to pitch-shifting mode and the original 'speed up = high voice' behavior is preserved.
Duration scales inversely with speed: a 60-minute podcast at 1.5× becomes 40 minutes; at 0.75× it becomes 80 minutes. Bitrate stays at 192 kbps, so file size scales the same way.
At 0.8×-1.5× the artifacts are inaudible on speech and minimal on music. Beyond that you may hear a subtle 'phasiness' on sustained tones - a tradeoff of any time-stretching algorithm, not specific to this tool.