Sample Rate, Bit Depth and the WAV File Format Explained

“44.1 kHz, 16-bit, stereo” is on every audio file’s spec sheet — but what do those numbers actually mean, and where does the file keep them? Understanding sample rate and bit depth (and the simple container that stores them) makes it obvious why a WAV is the size it is and how long it plays. Here’s the breakdown, plus how to read any file’s header with the WAV / AIFF Inspector.

Sample rate: how often the sound is measured

Digitising sound means measuring the waveform’s height many times per second. Each measurement is a sample, and the sample rate is how many you take per second, in hertz.

• 44.1 kHz (44,100 samples/sec) — the CD and streaming standard.
• 48 kHz — the video/film standard.
• 96 kHz / 192 kHz — high-resolution recording.

Why 44.1 kHz specifically? The Nyquist theorem says a sample rate can faithfully capture frequencies up to half its value — the Nyquist frequency. Human hearing tops out around 20 kHz, and 44.1 kHz / 2 = 22.05 kHz covers it with a little room for the anti-aliasing filter to roll off cleanly. The odd-looking 44,100 figure is a historical artefact: it fit neatly onto the video tape recorders used to master the first digital audio, and it stuck as the CD standard. Higher rates mainly help during production, not playback.

Here’s how the common rates line up with what they’re used for and the highest frequency each can represent:

Sample rate	Nyquist limit	Typical use
8 kHz	4 kHz	Telephone / voice
44.1 kHz	22.05 kHz	CD, streaming, music delivery
48 kHz	24 kHz	Video, film, broadcast
96 kHz	48 kHz	High-resolution recording
192 kHz	96 kHz	Studio mastering, archival

Recording above 48 kHz doesn’t make audible frequencies “sound better” on its own — the extra range sits above hearing — but it gives processing like pitch-shifting and steep filtering more room to work before artefacts creep into the audible band.

Bit depth: how precisely each sample is stored

Each sample’s height has to be rounded to a number, and bit depth is how many bits that number gets. More bits mean finer steps and a wider dynamic range (the gap between the quietest and loudest sound before noise or clipping):

• 16-bit → ~96 dB dynamic range (CD quality).
• 24-bit → ~144 dB, the extra headroom engineers want while recording and mixing.

Bit depth doesn’t change pitch or “loudness” — it changes precision. Each extra bit doubles the number of levels available and adds roughly 6 dB of dynamic range: 16 bits give 65,536 possible levels, while 24 bits give over 16 million. That deeper “floor” is why engineers record at 24-bit — you can set levels conservatively to avoid clipping and still keep the quiet detail well above the quantisation noise. The rule of thumb: record and mix at 24-bit, deliver at 16-bit.

A quick reference for the common depths:

Bit depth	Levels per sample	Approx. dynamic range	Where it’s used
8-bit	256	~48 dB	Legacy / retro, gritty samples
16-bit	65,536	~96 dB	CD, streaming delivery
24-bit	16,777,216	~144 dB	Recording, mixing, mastering
32-bit float	—	practically unclippable	DAW internal processing, field recorders

32-bit float is a special case: instead of fixed steps it stores each sample as a floating-point number, so levels that would “clip” in a fixed-point file can be pulled back down later without damage. It’s a working format, not a delivery one — you still export to 16- or 24-bit at the end.

Putting it together: file size and duration

Sample rate and bit depth directly determine the data rate of uncompressed audio:

bytes per second = sample rate × channels × (bit depth ÷ 8)

For 44.1 kHz, 16-bit, stereo that’s 44,100 × 2 × 2 = 176,400 bytes/sec (~1.4 Mbit/s). Divide the audio data size by that and you get the exact duration — no decoding required. It’s also why uncompressed audio is big: about 10 MB per minute at CD quality.

Worked example. Say a WAV’s data chunk is 30,870,000 bytes. First find the byte rate from its fmt chunk — 44.1 kHz, 16-bit, stereo gives 176,400 bytes/sec. Then 30,870,000 ÷ 176,400 = 175 seconds, or 2 minutes 55 seconds. No player, no decoder — just two numbers pulled straight from the header.

Switching the specs scales the size predictably. The same three-minute stereo recording weighs:

Specs	Byte rate	Size of a 3-minute file
44.1 kHz / 16-bit	176,400 B/s	~30.3 MB
48 kHz / 24-bit	288,000 B/s	~49.4 MB
96 kHz / 24-bit	576,000 B/s	~98.9 MB

Doubling the sample rate or bumping the bit depth from 16 to 24 doesn’t touch the audible content much — but it does grow the file in direct proportion, which is exactly why delivery formats settle back to 44.1 kHz / 16-bit.

Where the specs live: the WAV (RIFF) format

A WAV file is a RIFF container — a list of chunks. The layout is simple:

Chunk	Holds
RIFF / WAVE header	Marks the file as WAVE audio + total size
fmt	Sample rate, bit depth, channels, codec, byte rate
data	The raw PCM samples
LIST / INFO	Optional tags — title, artist, software

Everything on the spec sheet comes from the fmt chunk; the duration comes from the data chunk’s size. That’s the whole trick to reading a WAV’s metadata — you never touch the audio itself.

AIFF: the same idea, Apple-flavoured

AIFF (Apple’s format) is the same concept with two twists: it’s big-endian, and it stores the sample rate as an unusual 80-bit extended-precision float in its COMM chunk. Otherwise it carries the same facts — sample rate, bit depth, channels — and the duration comes from the number of sample frames divided by the sample rate.

Read a file’s specs instantly, in your browser

Because all of this lives in the header, you can read a file’s specs without decoding a single sample — which means it’s instant even on a huge file. The WAV / AIFF Inspector parses the RIFF/AIFF chunks entirely in your browser: drop in a .wav or .aiff and it shows the sample rate, bit depth, channels, codec, bitrate and exact duration, with the audio never leaving your device — handy for unreleased recordings.