Bitrate is the amount of data a recording uses per second, in kilobits per second (kbps). It decides both how big the file is and how much of the original sound survives.
Where the bits come from
A microphone produces a continuously changing voltage. To store it, a converter measures it many thousands of times a second (the sample rate) and rounds each measurement to one of a fixed set of values (set by the bit depth). Multiply the two, then by the number of channels, and you have the uncompressed bitrate.
Drag the sliders below. With few samples per cycle the stored steps can’t follow the wave’s fast wiggles; with few bits every step lands on a coarse grid, and the difference between the steps and the real wave is heard as noise.
Sampling, step by step
The cream line is the sound; the yellow steps are what gets stored. More samples follow fast wiggles (high frequencies); more bits make each step finer (less noise). Real audio uses 16 bits: 65,536 levels.
Now with real numbers
Uncompressed bitrate
1,411.2 kbps
WAV size per minute
10.1 MB
Highest frequency
22.05 kHz
Dynamic range
≈ 98 dB
Highest frequency is half the sample rate (the Nyquist limit). Human hearing tops out near 20 kHz, which is why CDs use 44.1 kHz. An MP3 at 128 kbps is 11× smaller than this.
For that same tactile approach elsewhere, ahaboo lets you drag the Moon around to see why it shows phases, and covers other topics built on trusted sources.
Compressed bitrate: MP3, AAC, Opus
CD audio runs at 1,411 kbps. A 128 kbps MP3 is 11 times smaller, so the encoder has to throw most of the data away. It does that with a psychoacoustic model: sounds masked by louder sounds nearby, and frequencies most adults can’t hear, get few or no bits. The bitrate you pick is the budget; the encoder decides how to spend it.
Bitrate
Size / min
Typical use
32–64 kbps
0.2–0.5 MB
Voice notes, audiobooks (mono). Opus handles this range best.
96–128 kbps
0.7–1 MB
Podcasts, casual music, streaming on mobile data.
160–192 kbps
1.2–1.4 MB
Music most people can’t tell from CD. AudioKit’s default.
256–320 kbps
1.9–2.4 MB
Maximum MP3 quality; DJ libraries, archiving lossy.
≈ 700–1,000 kbps
5–7 MB
FLAC (lossless, varies with the music).
1,411 kbps
10.1 MB
WAV / CD: 44.1 kHz, 16-bit, stereo.
Work out a file’s size
Size in bytes = bitrate in bits per second × seconds ÷ 8. The calculator does it for both compressed and uncompressed audio.
Audio file size calculator
≈ 4.8 MB(192 kbps × 210 s ÷ 8)
Choosing a bitrate
Speech: 64 kbps mono MP3 or 32–48 kbps Opus.
Music for listening: 192 kbps MP3 or 160 kbps AAC/M4A.
Anything you’ll edit again: WAV or FLAC, so the next export doesn’t stack losses.
Bitrate is how many bits of data are used for each second of audio, measured in kilobits per second (kbps). Higher bitrate means more detail kept and a larger file. File size ≈ bitrate × duration ÷ 8.
What is a good bitrate for MP3?
192 kbps is transparent for most people on most music. 128 kbps is fine for casual listening and speech. 320 kbps is the MP3 maximum. For podcasts, 64–96 kbps mono is standard.
Is 320 kbps better than 128 kbps?
It keeps more detail, especially cymbals, reverb tails and frequencies above 16 kHz. In blind tests many listeners can’t tell 192 from 320; more can tell 128 from 320 on good headphones with complex music.
What is the bitrate of a CD or WAV?
1,411.2 kbps: 44,100 samples per second × 16 bits × 2 channels. That’s about 10 MB per minute.
Does converting to a higher bitrate improve quality?
No. Detail discarded by the first encoder is gone. Re-encoding a 128 kbps MP3 at 320 kbps gives a bigger file that sounds the same or very slightly worse.
What is the difference between CBR and VBR?
Constant bitrate (CBR) uses the same number of bits every second. Variable bitrate (VBR) spends more on complex passages and fewer on simple ones, giving better quality for the same average size. Both play everywhere today.