Audio Compressor
Shrink audio files by re-encoding to a smaller MP3 bitrate, locally in your browser.
Loading Audio Compressor… If nothing happens, please enable JavaScript.
An audio compressor reduces the file size of a sound recording by lowering its bitrate, the amount of data used per second of audio. A voice memo, podcast, lecture or song that is too large to email or upload becomes far smaller while still sounding fine, because most recordings are saved at a higher bitrate than they actually need. This tool re-encodes your audio to MP3 at the bitrate you choose, trading a little fidelity for a much smaller file.
Frequently asked questions
Is my audio uploaded anywhere?
How much smaller will my file get?
What bitrate should I choose?
Will compressing reduce the quality?
Why does it output MP3?
Can I compress a file that is already small or low quality?
Is there a file size limit?
Does it work offline?
About Audio Compressor
Everything happens inside your browser using a WebAssembly build of FFmpeg. Your file is read from your device, compressed locally, and returned as a download. Nothing is ever uploaded, so it is safe for private voice notes, interviews or confidential recordings, and there are no upload-size limits beyond your device's memory.
Bitrate is the dial that balances size against quality. 192 kbps is near-transparent for music; 128 kbps is the classic standard that sounds good for most listening; 96 kbps noticeably shrinks files with modest quality loss; and 64 kbps is ideal for spoken-word content like podcasts and audiobooks, where the smaller size matters more than musical detail. Lower bitrate always means a smaller file, so pick the lowest setting that still sounds acceptable for your use.
A caveat worth knowing: compression is lossy and cumulative. Compressing an already low-bitrate file, or compressing the same clip repeatedly, degrades quality each time, so always start from the highest-quality source you have. Because the tool runs single-threaded in the browser to stay private and compatible, very long recordings take longer and use more memory than a desktop encoder, but typical files compress quickly.
What a 'kilobit per second' actually buys you
Bitrate, measured in kilobits per second (kbps), is simply how much data the encoder spends on each second of sound. A 128 kbps MP3 uses 128,000 bits, about 16 kilobytes, for every second of audio. Multiply that out and a three-minute song lands around 2.9 megabytes, which is why bitrate is the single biggest factor in an audio file's size.
The interesting part is how little we actually need. Decades of psychoacoustic research show that human hearing is easily fooled: loud sounds mask quieter ones nearby, and we are far more sensitive to some frequencies than others. Encoders exploit this, spending bits where the ear is sharp and saving them where it is not. That is why a well-encoded 128 kbps file can sound almost identical to a CD that uses more than ten times the data.
Speech needs even less. The human voice occupies a narrow frequency range with predictable structure, so codecs designed for it can stay perfectly intelligible at 32 to 64 kbps, which is how phone calls and podcasts stream comfortably on weak connections. It is the same principle that lets a compressor shrink a spoken-word recording to a fraction of its size while leaving every word clear.