Compress Audio
Make an audio file small enough to send.
Using toolbook means you agree to our Terms and Privacy Policy.
or drag them here
MP3, M4A, WAV, FLAC, or OGG · as many as you like
- Takes
- MP3, M4A, AAC, WAV, FLAC, OGG, and Opus. You always get an MP3. Do several at once with the same settings. AIFF and WMA don’t work in a browser.
- Size
- No limit. A 29 MB, 12-minute MP3 took about four seconds here.
Your files stay on your device
How to use it
Drop an audio file here and pick a lower bitrate to make it smaller. You see the new size before you start. Your file stays on your device.
Drop your audio file onto the box above. You’ll see its current bitrate — that’s how much data it uses per second of sound. Pick a lower one and the new size appears before you start.
For voice recordings like lectures, interviews, or voice memos, 64 kbps mono sounds perfectly clear and is tiny. For music, stay at 128 kbps or above, or cymbals and applause start to sound fuzzy.
Going higher than your file already is won’t help — it makes a bigger file that sounds the same (slightly worse, if anything). The tool warns you if you try.
The result is always an MP3, which plays on every device. Compressing loses some quality permanently, so keep the original if it matters.
FAQ
How small can I make an audio file?
At 64 kbps mono — great for speech — an hour is about 28 MB. At 128 kbps stereo it’s about 60 MB, and as a WAV about 600 MB.
Why did my file get bigger?
You picked a higher bitrate than the file already had. Pick a lower one.
Does compressing lose quality?
Yes, permanently. If the recording matters, compress a copy.
Should I use mono?
For speech, yes — it sounds better at small sizes. For music, no.
Which formats work?
MP3, M4A, AAC, WAV, FLAC, and OGG. You always get an MP3 back.
How it worksFor the curious
Bitrate is bits spent per second of sound. A 15-minute recording at 256 kbps is 29 MB; at 96 kbps it is 11 MB. A constant-bitrate MP3 spends the same number of bits every second, so the expected size is length × bitrate — arithmetic rather than an estimate — and the finished file lands within about a percent of it.
The encoder discards what is hardest to hear. Mono halves the information to carry, so 64 kbps mono sounds better than 64 kbps stereo on a voice. Discarded detail is not stored anywhere, so re-encoding at a higher bitrate only writes more bits about the same audio.
Decoding uses the codecs built into your browser; encoding uses a WebAssembly build of the LAME MP3 encoder, loaded before processing starts.