No upload, 100% local, no account

Audio pitch

Make a voice deeper or higher without speeding it up or slowing it down. Everything runs in your browser; your file never leaves your device.

How Audio pitch works

Audio pitch shifts a track up or down by semitones, from one octave down to one octave up, without changing its duration, entirely inside your browser. The shipped ffmpeg.wasm build has no dedicated high-quality pitch shifter compiled in, so this tool builds one from filters that ARE always available: it first normalises the audio to a known 48000 Hz sample rate (aresample), then relabels that rate by the pitch ratio (asetrate), which shifts both pitch and speed together, and finally compensates the speed change with the same atempo time-stretch chain audio-speed uses, restoring the original duration while keeping the new pitch.

This is a real, honest trade-off: it is a classic sample-rate based shifter, not a studio-grade phase-vocoder (like Melodyne or a rubberband-based tool), so large shifts can sound slightly artificial on complex material such as a full mix, while smaller shifts on a single voice or instrument tend to sound clean.

Written by Bastien Sulyan

How to use Audio pitch, step by step

  1. Drop or select the audio file you want to shift.
  2. Set the pitch shift in semitones with the slider or a preset (from -12 to +12).
  3. Click Start. ffmpeg resamples and time-stretches the file locally.
  4. Preview the result and download it once processing finishes.

Common use cases

  • Lowering a voice recording by a few semitones for a deeper, more serious tone without slowing the speech down.
  • Raising a sample or vocal snippet to match the key of a backing track.
  • Creating a simple voice-disguise effect for a video or podcast segment.
  • Testing how a melody sounds transposed to a different key before committing to a full re-recording.

Frequently asked questions

Is my audio uploaded anywhere to shift its pitch?

No. ffmpeg compiled to WebAssembly runs the resample/asetrate/atempo chain inside your browser tab. Your file is read into memory on your device, processed there, and the shifted result is offered as a local download. Open your browser DevTools Network tab while running the tool: no request carries your audio.

Does the duration change when I shift the pitch?

No, that is the whole point of the tool. The sample-rate trick that shifts pitch also changes speed as a side effect, so the atempo filter chain compensates by the inverse factor, restoring the original duration while keeping only the pitch shift.

Why does the tool normalise to 48000 Hz first?

The asetrate trick needs to know the exact current sample rate to compute a correct new one; probing the real sample rate of an arbitrary input file adds real complexity. Normalising every input to a known 48000 Hz first with aresample means the pitch ratio applied afterwards is always accurate, regardless of what sample rate the source file actually used.

How does this compare to a professional pitch-correction tool?

It is a real pitch shift, not a gimmick, but it uses the classic asetrate+atempo combination rather than a phase-vocoder algorithm. Studio tools built around phase-vocoder or PSOLA techniques generally sound cleaner on large shifts and complex mixes; this tool is well suited to moderate shifts on a single voice or instrument, and to quick, no-install experiments.

Does the output keep the same audio format as the input?

Yes. The file is processed and returned in the same container and codec as the input, matching audio-speed. At zero semitones (no shift), the tool takes a lossless shortcut and simply copies the stream without re-encoding.

Does audio pitch work offline?

Mostly, yes. The pitch shift itself always runs on your device, so your audio never leaves your machine. The one caveat: ffmpeg's engine is a sizeable download fetched from this site the first time you use any ffmpeg-based tool. Once that download and the page have both completed, you can disconnect from the network and the pitch shift still works.