Compress Audio
100% private — runs on your device, never uploaded. Works offline once loaded.
Two ways to make an audio file smaller, and the second is the one people actually need. Compressing by quality lets you pick a bitrate — 128 kbps for music, 96 or 64 for speech — and takes whatever size that gives. Compressing to a target size works backwards from the limit you have been handed: give it 8 MB and it works out the bitrate that fits your recording into 8 MB and encodes once at that rate. For speech, switching to mono halves the size again at no real cost, because a voice recorded on one microphone carries no stereo information worth keeping.
What bitrate really costs you
Bitrate is how many bits a second the encoder may spend. Speech needs very few: a voice recorded on one microphone is transparent at 64 kbps in mono, which is about 28 MB an hour — small enough to email an entire interview. Music needs more, because cymbals, reverb and stereo detail are exactly what the encoder throws away first. Below roughly 128 kbps most listeners start to hear it on decent headphones.
Mono is the other half of the saving, and it is free for speech. One microphone captured one signal; storing it as two identical channels doubles the file for nothing. It is the wrong move for music, where the stereo image is part of the recording.
When compression cannot help
If the file is already at a low bitrate, re-encoding it lower loses quality fast and saves little — the encoder is now discarding detail from audio that has already been stripped. If you need a big reduction from an already-compressed file, cutting the length or switching to mono usually beats dropping the bitrate again.
And if the file is already under your target, this tool refuses rather than re-encoding. Handing back a file that is the same size but worse would be a bad trade quietly made on your behalf.
Frequently asked questions
What bitrate for a podcast?
96 kbps mono is the common standard for spoken word — about 43 MB an hour and indistinguishable from higher rates on a voice. 128 kbps if there is music in the intro.
Will compressing fix a quiet or noisy recording?
No — those are different jobs. Use Normalize Audio for levels and Remove Background Noise for hiss and hum, then compress at the end.
Advertisement