MP3 bitrate explained
What each bitrate actually sounds like, the sizes they produce, and why speech and music want completely different numbers.
In short
Bitrate is how much data per second an MP3 spends. 128 kbps is the everyday default and roughly 1MB per minute. Speech is fine at 64–96 kbps; music generally wants 192 kbps or more.
An MP3's bitrate is how many kilobits per second it spends describing the sound. It is the only quality control most MP3 encoders expose, and it maps almost exactly onto file size.
Size is easy to predict
Megabytes per minute is roughly bitrate ÷ 133. So 128 kbps is about 0.96MB per minute — near enough one megabyte a minute, which is a useful number to carry around.
| Bitrate | Per minute | Per hour | Suits |
|---|---|---|---|
| 64 kbps | ~0.5 MB | ~29 MB | Voice notes, lectures, audiobooks |
| 96 kbps | ~0.7 MB | ~43 MB | Podcasts and interviews — generous for speech |
| 128 kbps | ~1.0 MB | ~58 MB | The general-purpose default |
| 192 kbps | ~1.4 MB | ~86 MB | Music where quality matters |
| 256 kbps | ~1.9 MB | ~115 MB | Music, close to transparent for most listeners |
What each range sounds like
Below about 96 kbps, music noticeably thins out — cymbals and sibilance take on a watery, swirling quality as the encoder runs short of bits for high frequencies. Speech survives this far more gracefully, because a voice occupies a narrow frequency range and there is simply less to describe.
Between 128 and 192 kbps, most people stop being able to identify the compressed version reliably on normal equipment. Above 256 kbps you are into diminishing returns for almost everyone; if genuine fidelity matters, the answer is a lossless format rather than a very high MP3 bitrate.
Mono halves it again
A single voice has no meaningful stereo content, and many devices play only one channel anyway. Converting a spoken-word recording to mono roughly halves the file on top of whatever the bitrate saves — often the largest single saving available on a podcast or voice note.
Mistakes worth avoiding
- Raising the bitrate of an existing MP3. Re-encoding a 128 kbps file at 256 kbps produces a bigger file with the same damage, not a better one.
- Using 320 kbps for speech. It is roughly three times the size of 96 kbps for no audible benefit on a voice.
- Compressing from an already-compressed source when the original WAV is available. Always encode once, from the best source you have.