Compress Audio
Processed locally in your browser — nothing is uploaded.
Quick answer
The simplest way to shrink an audio file is to lower its bitrate — the kilobits per second used to store the sound. A file’s size is roughly its bitrate multiplied by its duration, so halving the bitrate roughly halves the size. This tool lets you pick a bitrate and shows a live size preview before you export, all in your browser with nothing uploaded. First use downloads the ~32 MB engine once.
How to compress audio
Add your audio
Drop in the file you want to shrink. Its current bitrate and size are shown so you know your starting point.
Choose a target bitrate
Slide the bitrate down and watch the live estimated output size update. Lower bitrate means a smaller file and gradually softer quality.
Compress and download
Export at your chosen bitrate and download the smaller file.
How does bitrate decide the file size?
Bitrate is how many kilobits are spent on each second of audio. The maths is almost that simple: file size ≈ bitrate × duration. A three-minute song is 180 seconds; at 320 kbps that is about 7.2 MB, and at 128 kbps about 2.9 MB — the same music, 60% smaller, because each second is allotted fewer bits. That is why the size preview can update the instant you move the slider: it does not need to encode the file to know how big it will be, only to multiply.
| Bitrate | Approx. size | Sounds like | Best for |
|---|---|---|---|
| 320 kbps | ≈ 7.2 MB | Transparent | Music you want pristine |
| 192 kbps | ≈ 4.3 MB | Very good | General music listening |
| 128 kbps | ≈ 2.9 MB | Good | Web streaming, email |
| 96 kbps | ≈ 2.2 MB | Acceptable | Spoken word, podcasts |
| 64 kbps | ≈ 1.4 MB | Noticeably softer | Voice memos, tight limits |
Compression is lossy: the bits removed to hit a lower bitrate are gone, and re-compressing a file that is already low-bitrate makes it worse, not smaller-for-free. Start from the highest-quality source you have.
One nuance the simple formula hides is the difference between constant and variable bitrate. A constant bitrate (CBR) spends the same bits on every second, which makes the final size trivially predictable — that is the number this tool previews. A variable bitrate (VBR) spends more on complex passages and less on simple ones, often sounding better at the same average size, but turning the final size into an estimate rather than a promise. When you must land under a hard email limit, CBR’s predictability is the safer choice.
Should speech and music use the same bitrate?
No. Speech is far more forgiving than music: a podcast at 96 kbps mono sounds clean, while the same bitrate on a dense mix sounds smeared. If your file is a voice recording, dropping to mono roughly halves the size again on top of the bitrate cut, because you are storing one channel instead of two — see mono to stereo for when that trade-off is safe.
Sample rate is a second lever people forget. A voice captured at 48 kHz carries high frequencies no one needs for dialogue, so downsampling to 44.1 kHz — or even 32 kHz for pure speech — trims size with no audible cost. Music, which uses the full frequency range, should stay at 44.1 kHz or higher, where dropping the rate would dull the top end.
Does compressing upload my file?
No. The re-encode is done by ffmpeg.wasm in your browser; your audio is read, compressed and returned as a download without ever leaving your device. Open DevTools → Network and compress a file — you will see no upload. That is the whole point: you can shrink a confidential recording for email without handing it to a third-party server first.
Related audio and video tools
If loudness varies wildly across the file, normalize it before compressing for a cleaner result at low bitrates. To change format as well as size, use the audio converter. Trimming dead air with trim audio shrinks the file for free by cutting duration. And if it is really a video you need smaller, compress a video instead.
Your files never leave your device
Processed locally in your browser — nothing is uploaded. You can verify it yourself: open your browser’s developer tools, watch the Network tab, and use the tool — no upload request is ever made.
Frequently asked questions
What bitrate should I use to email an audio file?
For speech, 96 kbps mono is small and clear. For music, 128 kbps is a good size-to-quality balance. Use the live size preview to land under your email limit.
Does lowering the bitrate lose quality?
Yes. Compression is lossy — fewer bits per second means less detail. The loss is subtle at 192–320 kbps and more audible below about 96 kbps.
Why is my file barely smaller after compressing?
It was probably already low-bitrate. You cannot reclaim much from a file that is already compressed; start from a higher-quality source.
How is the output size known before I export?
Because size is essentially bitrate times duration. Multiplying the two gives an accurate estimate without having to encode first.
Can I compress a WAV?
Yes, and the saving is dramatic — WAV is uncompressed, so encoding to a 128 kbps MP3 can cut it by 90% or more.
Will mono make it smaller?
Yes, roughly half again, because you store one channel instead of two. It is ideal for voice, less so for music where stereo matters.
Is my audio uploaded to compress it?
No. Everything runs locally in your browser via WebAssembly. You can verify there is no upload in the Network tab.