Audio Bitrate vs. Sample Rate: What's the Difference?

Bitrate and sample rate are the two specifications people mix up most often when discussing audio quality. They measure completely different things, they apply to different format families, and confusing them leads directly to wrong format-and-settings decisions. This is the explainer that untangles them.

Sample Rate: How Often the Waveform Is Measured

Sample rate is the number of audio measurements taken per second, expressed in Hertz (Hz) or kilohertz (kHz). At 44,100 Hz (44.1 kHz), the audio interface captures the instantaneous voltage of the analog signal 44,100 times every second and writes each measurement as a number.

The Nyquist-Shannon sampling theorem says you need to sample at more than 2× the highest frequency you want to reproduce. Human hearing stops around 20 kHz, so 44.1 kHz captures everything up to 22.05 kHz with a small guard band — this is why CD audio uses 44.1 kHz, and why nearly every consumer audio file you have ever played is at that rate.

Common sample rates and where they're used:

8 kHz — telephone, AM radio. Up to 4 kHz audio.
22.05 kHz — old web audio, low-quality voice memos.
44.1 kHz — CD, consumer music, most podcasts.
48 kHz — video, broadcast, DAW default. Standard for film, TV, and YouTube.
88.2 / 96 kHz — high-resolution music masters.
176.4 / 192 kHz — studio masters with processing headroom.

Higher sample rates do not improve audio you can hear; they capture frequencies above human hearing, which matters only as headroom during DSP processing in studios.

Bit Depth: How Precisely Each Sample Is Stored

Closely related and easy to confuse with sample rate. Bit depth is the number of bits used to represent each sample's amplitude:

16-bit: 65,536 possible amplitude values. 96 dB of dynamic range. CD standard.
24-bit: 16,777,216 values. 144 dB of dynamic range. Pro audio standard.
32-bit float: effectively unlimited dynamic range; used inside DAWs for processing.

Bit depth governs the noise floor and dynamic range. It does not change how the audio sounds at normal listening levels — it changes how much processing headroom is available before quantization noise becomes audible.

Bitrate: Data Rate of the Encoded File

Bitrate applies to compressed (usually lossy) formats: MP3, AAC, Opus, Vorbis. It is the number of bits per second the encoded file uses. A 192 kbps MP3 uses 192,000 bits per second of audio playback.

Bitrate is not a direct quality measure across codecs. Codec efficiency varies dramatically:

96 kbps Opus is roughly equivalent to 192 kbps MP3 in perceived quality.
128 kbps AAC is roughly equivalent to 160-192 kbps MP3.
64 kbps HE-AAC is comparable to 128 kbps MP3 for voice.

Comparing bitrates only makes sense within a single codec. The ladder is meaningful for "should this MP3 be 128 or 256 kbps," not for "is 192 kbps MP3 better or worse than 192 kbps AAC."

How They Multiply for Uncompressed Audio

For lossless PCM (WAV, AIFF, the audio inside FLAC), the data rate is fully determined by sample rate × bit depth × channel count. CD audio:

44,100 samples/sec × 16 bits/sample × 2 channels = 1,411,200 bits/sec = 1,411 kbps

A 24-bit/96 kHz stereo studio master:

96,000 × 24 × 2 = 4,608,000 bits/sec = 4,608 kbps

This is the "uncompressed bitrate." Lossless codecs like FLAC compress this number to about 50-60% of the raw rate without changing what plays back. Lossy codecs throw away data the encoder judges inaudible, dropping the rate to 96-320 kbps depending on quality target.

How Lossy Codecs Decouple Sample Rate from Bitrate

This is where confusion peaks. With uncompressed PCM, raising sample rate raises file size proportionally. With lossy codecs, you can have a high sample rate at a low bitrate — the encoder allocates its bit budget across the higher number of samples, sacrificing some precision per sample to maintain the temporal resolution.

A 48 kHz / 96 kbps Opus file is a perfectly normal thing. The encoder analyzes 48,000 samples per second per channel and represents them in 96,000 bits per second using psychoacoustic modeling. The audio is lossy but not noticeably worse than the same content at 44.1 kHz / 96 kbps.

This decoupling is why "what's the bitrate?" and "what's the sample rate?" are independent questions for compressed audio.

The Practical Cheat Sheet

When choosing settings:

For WAV or FLAC (lossless): pick sample rate and bit depth. There is no bitrate setting. The data rate is determined by your choices.
For MP3 or AAC (lossy): pick bitrate. Sample rate is usually 44.1 kHz or 48 kHz; the encoder will accept others.
For Opus (lossy): pick bitrate. Internally Opus operates at 48 kHz; lower-rate inputs are upsampled.

Recommended settings by use case:

Music production master: 48 kHz, 24-bit WAV.
Consumer music distribution: 44.1 kHz, 16-bit FLAC for archival; 192-256 kbps AAC for streaming; 320 kbps MP3 for legacy compatibility.
Podcast: 44.1 kHz, 96-128 kbps mono MP3 for delivery; 48 kHz, 24-bit WAV for the source recording.
Video soundtrack: 48 kHz, 24-bit WAV throughout production. AAC at 192 kbps for delivery.
Web audio for game or app: 48 kHz Opus at 96 kbps for music, 32-48 kbps for voice.

Common Confusions

"My 192 kHz file must sound better than my 48 kHz file." Probably not. Above 48 kHz, the extra data is inaudible to humans; you are storing ultrasonic frequencies for processing headroom that does not exist on consumer playback.
"My 320 kbps MP3 must sound better than my FLAC." No. FLAC is lossless and bit-perfect; 320 kbps MP3 is the best lossy MP3 can deliver but is still a lossy reduction.
"WAV at 128 kbps." Does not exist. WAV stores PCM at the sample rate × bit depth × channels rate. You cannot set a "bitrate" for WAV.
"What's the bitrate of my FLAC?" FLAC files have a calculated effective bitrate based on compression ratio and source PCM rate, but it is not a setting. Compression level changes file size, not data integrity.

For deeper coverage of either axis, see the audio bitrate guide, the sample rate explainer, and the bit depth deep dive. The lossless vs lossy primer covers when each format family is the right choice.