A shift of n semitones in the Audio Pitch Changer uses a playback rate of 2^(n/12), so raising a track by +12 semitones doubles its playback speed while cutting the duration roughly in half. GarageBand, by contrast, exposes its pitch controls through the Region inspector and a Transpose command on the Track menu, both of which act on a selected region inside a project. That makes GarageBand's approach well suited to retuning a vocal inside a session, but it requires the project to be open and the source to live inside a track. When the goal is to step an external audio file up or down by a clean octave and walk away with a new file, a browser-based local tool often fits the task better. The Audio Pitch Changer accepts whole-number semitone shifts from -12 through +12, renders the result in a single offline pass, and exports a fresh PCM16 WAV file entirely within the current tab.
GarageBand's Built-In Pitch Control on Mac
GarageBand on Mac exposes two related pitch controls that act on a selected region: a Pitch parameter measured in cents inside the Region inspector, and a Transpose command on the Track menu that shifts the region by whole semitones. Together they cover small tuning corrections and larger interval moves without forcing you to reopen the project in a full DAW. The pitch controls live at the region level, so they only affect the audio you select and leave other regions untouched.
- Open the GarageBand project that contains the audio region you want to retune.
- Click the region once to select it. The Region inspector appears on the right side of the window.
- In the Region inspector, locate the Pitch field and type a value in cents. One semitone equals 100 cents, per the standard twelve-tone equal-tempered scale documented by McGill University course material.
- For semitone-sized moves, choose Track > Transpose (or right-click the region and choose Transpose), then enter a whole number of semitones, positive or negative.
- Press Play to audition the change against the rest of the project.
- If you need the changed audio as a separate file, bounce the region: choose Share > Export Song to Disk, pick a destination, and select an audio format such as AIFF or WAV.
GarageBand's settings also include Audio/MIDI preferences for input and output devices, song-level controls for tempo, key, and time signature, and Advanced settings that govern recording bit depth and export volume. None of those are required for a single pitch move, but they explain why GarageBand treats pitch as a region-level parameter rather than a file-level one. The trade-off is that you cannot move a non-region file at all until you import it into a track first.
When a Browser Tool Fits Better
GarageBand is excellent when the file already lives inside an open session and the pitch move is part of a larger arrangement. The workflow becomes awkward when you only need the changed audio as a fresh standalone file, when you are on a device that does not run GarageBand, or when you want a one-shot octave shift without managing tracks and regions. A browser-based tool that processes a single uploaded file locally and hands back a downloadable WAV removes all of those steps at once, and it keeps the file out of any project library so the original stays untouched. For readers working purely in a browser, a browser-focused walkthrough covers the broader pattern with the same underlying engine.
How to Change Pitch with the Audio Pitch Changer
The Audio Pitch Changer runs entirely in the current tab: no upload, no account, no plugin install. The whole workflow fits in three choices and one download.
- Choose one browser-decodable audio file no larger than 50 MiB. Common MP3, WAV, M4A, AAC, Ogg, WebM, and FLAC inputs are accepted, although actual codec support still depends on the browser.
- Set a whole-number pitch shift from -12 through +12 semitones. A setting of +12 raises the material by one octave, -12 lowers it by one octave, and 0 leaves the pitch untouched. The duration will change in the opposite direction: raising the pitch shortens the audio, while lowering the pitch lengthens it.
- Create the complete local render. The tool reports the new duration before you download so you can confirm the file size is what you expected.
- Download the result as an uncompressed PCM16 WAV. The export does not preserve the original codec, tags, artwork, or container metadata, so treat the output as a fresh file rather than a re-encoded copy.
Pitch and Duration Are Linked: What Actually Changes
This implementation uses playback-rate resampling, which means the playback rate, pitch, and duration all move together. The rate for a shift of n semitones is exactly 2^(n/12), and MDN documents the matching relationship: a non-unit playbackRate causes resampling, with values below one slowing the audio and values above one speeding it up. W3C's Web Audio specification defines the same cents relationship in the AudioBufferSourceNode playback algorithm, where detuning contributes a factor of 2^(detune/1200).
Worked example: a 60-second clip shifted by +12 semitones uses a playback rate of 2^(12/12) = 2. The rendered file is therefore roughly 60 / 2 = 30 seconds long, and the perceived pitch sits one octave above the original. The same logic applies in reverse: a -12 shift uses a 0.5× rate and produces roughly twice the original duration. The tool does not claim to preserve tempo, speech cadence, beat positions, or exact original duration. If you need pitch change at fixed duration, a phase-vocoder or other time-stretching workflow is the right tool.
Limits, Formats, and Output Specs
The tool checks its limits before it does any expensive rendering work, so a rejection at the validation stage is cheap and a download always represents a fully rendered file. The following table lists the exact endpoint values defined by the equal-tempered formula and the resource ceilings that gate a successful render.
| Setting | Exact playback rate | Approximate duration effect |
|---|---|---|
| -12 semitones (one octave down) | 0.5× | Roughly 2× the original duration |
| 0 semitones | 1× | Original pitch and duration |
| +12 semitones (one octave up) | 2× | Roughly 0.5× the original duration |
Beyond the shift range, the tool enforces a set of hard ceilings. One encoded input may be at most 50 MiB. Decoded audio is capped at five minutes, eight channels, and a sample rate between 8,000 Hz and 192,000 Hz. Input decoding is limited to 30 million channel samples, and the pitch-shifted output is separately limited to 30 million channel samples. Because lowering pitch lengthens the file, an input that passes the decoded limit can still fail the output limit; in that case the tool reports that the requested shift is over budget and produces no download. Empty files, unsupported types, invalid decoded dimensions, and failed decodes are all rejected up front, before any audio is rendered.
Practical Use Cases for Whole-Octave Shifts
Whole-octave shifts are useful in a handful of predictable situations. Sound designers reach for them when prototyping chip-tune-style arpeggios, thickening a bass patch, or generating call-and-response variations from a single source take. Music students use them to build practice tracks that sit above or below their comfortable range without recording new material. Voice-memo authors sometimes raise the pitch of a long dictation to skim it faster, since the linked speed-up reduces listening time alongside the audio. In all of these cases the trade-off is the same: pitch and duration move together, so the workflow is creative or experimental rather than transparent mastering. For final masters, or any task that must keep duration fixed while changing pitch, a dedicated phase-vocoder or elastic-audio workflow remains the right choice.
The downloadable file is always a freshly encoded PCM16 WAV, so treat the result as a starting point rather than a lossless rewrap of the source. Original codec compression, ID3 tags, embedded artwork, chapter markers, and other container metadata are not carried over; if you need that metadata preserved, reattach it to the WAV in your own library afterwards. Strong upward shifts can also reveal aliasing or codec artifacts that were inaudible in the original, since the browser's resampler is implementation-defined and the source may carry hidden artifacts of its own. Processing and sample data remain in the current browser tab throughout, so the same file processed locally on a different device or browser can still sound subtly different even with identical settings.