How MP3 Decides What Sound to Throw Away
In a hurry? Skip straight to the numbers.
Open the Audio File Size Calculator →This calculator's compressed-format formula treats MP3 file size as simply bitrate times duration - clean arithmetic that hides a genuinely clever piece of applied psychoacoustic science determining exactly which parts of the original audio survive that compression and which parts get discarded entirely.
Why Simply Removing Random Data Wouldn't Work
Shrinking an uncompressed audio file down to a much smaller lossy file requires discarding real information - there's no way around that basic fact. But MP3 compression doesn't discard data randomly or uniformly; it specifically targets the parts of the original sound that human hearing is least likely to actually notice are missing, exploiting known, well-researched limitations and quirks of human auditory perception rather than simply degrading the entire signal evenly.
Psychoacoustic Masking: The Key Principle Being Exploited
Human hearing has a well-documented masking effect - a loud sound at one frequency makes it genuinely harder, or in some cases impossible, to perceive a quieter sound at a nearby frequency occurring at the same time, even though both sounds are physically present in the actual audio signal. MP3 encoders analyze the audio in small time segments, model which quieter frequency components would likely be masked (rendered inaudible) by louder nearby components at that same moment, and then allocate fewer bits - or none at all - to encoding those specific masked components, since removing information the ear couldn't have perceived anyway produces essentially free file size reduction with no audible cost.
The Fraunhofer Institute's Research Behind the Format
The MP3 format's development is closely associated with research conducted at the Fraunhofer Institute for Integrated Circuits in Germany, beginning in the 1980s, where researchers including Karlheinz Brandenburg worked for years refining exactly this kind of psychoacoustic modeling into a practical, standardized compression algorithm - work that required extensive listening tests and iterative refinement to correctly calibrate which specific data could be discarded without most listeners noticing a meaningful quality loss, eventually leading to the format's formal standardization in the early 1990s as part of the broader MPEG video and audio standards effort.
| Compression type | What it removes | Typical size reduction |
|---|---|---|
| Lossless (e.g. FLAC) | Only statistical redundancy - fully reversible, no perceptual assumptions | Roughly 40-60% smaller than uncompressed |
| Lossy (e.g. MP3) | Both redundancy and perceptually-masked information, deliberately irreversible | Often 80-90%+ smaller than uncompressed |
This is exactly why lossy formats can achieve such dramatically smaller file sizes than lossless compression alone could ever manage - lossless methods are fundamentally limited to removing only genuinely redundant data that can be perfectly reconstructed, while lossy psychoacoustic compression additionally removes real audio information, betting specifically (and, when well-implemented, quite successfully) that human hearing won't notice its absence.
Applying This to a Calculated Compressed File Size
The bitrate figure driving this calculator's compressed-format formula represents a deliberately chosen tradeoff point - lower bitrates force the psychoacoustic model to discard more information, eventually reaching a point where audible artifacts appear because even perceptually-relevant information has to be sacrificed to hit the target size, exactly why streaming services and audio professionals treat bitrate choice as a genuine quality decision rather than a purely technical file-size setting.
Ready to Put This Into Practice?
Now that you understand how it works, plug in your own numbers and get an instant, accurate result.
Use the Audio File Size Calculator Now →