How Bitrate Actually Affects File Size and Sound

What bitrate is

Bitrate (measured in kilobits per second, or kbps) is the number of bits of audio data stored per second of playback. It's a rate — how much data is used to represent each second of the recording.

In a lossy format like MP3, bitrate controls how aggressively the encoder discards audio data. A higher bitrate means less is discarded; a lower bitrate means more is discarded. The encoder is always making trade-offs: which frequencies to keep, which to approximate, which to remove. Bitrate is the budget it has to work with.

The file size calculation

File size from bitrate is straightforward:

For a 3-minute (180-second) song at 192 kbps:

At 320 kbps, the same song would be approximately 7.2 MB. At 128 kbps, it would be approximately 2.9 MB. The relationship is linear — doubling the bitrate doubles the file size.

Approximate. Variable bitrate encoding can produce smaller files than constant bitrate at the same nominal setting.

What actually degrades at low bitrates

When the encoder has too little bitrate budget, it has to make harder trade-offs. The first things to go are the subtle details: ambience, room sound, very high and low frequencies, the fine texture of instruments. With less budget, the encoder also does a rougher job on fast transients — drums, percussion, sharp attacks — which can produce a pre-ringing artefact (a faint "smear" before the actual sound).

At 128 kbps, most people can detect quality loss on complex material — busy mixes, acoustic instruments, cymbals. The sound can feel slightly closed-in or have a muffled quality in the upper frequencies.

At 96 kbps and below, artefacts are consistently audible: "watery" or "bubbly" distortion on sustained notes, obvious frequency roll-off, and a generally compressed feel to the mix.

At 64 kbps — used for voice-optimised encoding, podcasts at low bandwidth, or older streaming services — the encoding is very aggressive. Speech remains intelligible but music quality degrades substantially.

The transparency threshold

In audio, "transparent" means the listener cannot reliably tell the lossy file apart from the lossless original under controlled listening conditions. For MP3, the transparency threshold is generally considered to be around 192 kbps.

At 192 kbps, a well-encoded MP3 is indistinguishable from WAV for most people on typical headphones and speakers. This doesn't mean 192 kbps is identical to WAV — it means the difference is below the threshold of consistent human perception in normal listening conditions.

The encoder matters too. Different MP3 encoders at the same bitrate produce different results. LAME (the open-source MP3 encoder) at 192 kbps V0 (variable bitrate) is considered among the highest-quality MP3 encodings available and is effectively transparent for nearly all listeners. Most conversion tools, including server-side ones, use LAME or an equivalent.

When 320 kbps makes sense

320 kbps is the maximum for MP3. The step from 192 to 320 produces files about 65% larger. The audible difference for most people on consumer equipment is minimal to nonexistent.

There are cases where 320 kbps is worth it: archiving a lossy-only collection (when you don't have lossless originals and want to maximize what you keep), encoding for DJ software or monitoring situations where artefacts at 192 kbps are occasionally detectable, or simply as a personal preference when storage is not a constraint. These are edge cases — most people encoding for personal use at 192 kbps will not benefit from switching to 320.

Which to choose

• 128 kbpsVoice-only content (podcasts, audiobooks), bandwidth-constrained distribution, small file size with acceptable quality for speech
• 192 kbpsGeneral music, sharing, streaming, podcasts with music. The practical default — transparent for most uses
• 256 kbpsHigher confidence margin over 192 kbps for mixed content. Used by Apple Music as their AAC delivery bitrate
• 320 kbpsMaximum MP3 quality. Minimal audible gain over 192 kbps for most listeners. Useful when storage is not a concern or you want headroom for further processing

Variable vs constant bitrate

Constant bitrate (CBR) uses the same bitrate for every second of audio, regardless of the complexity of the content. Simple sections (silence, steady tones) get the same budget as complex sections (dense mixes, fast transients).

Variable bitrate (VBR) allocates more bits to complex sections and fewer to simple ones. At the same average bitrate, VBR generally produces better sound quality than CBR, because the encoder uses its budget where it's most needed.

For most conversion purposes — sharing, streaming, personal use — VBR at an equivalent quality setting to 192 kbps CBR is the better choice. All modern players support VBR MP3. The resulting file is often slightly smaller than the equivalent CBR file while matching or exceeding its quality.