Audio formats explained
Six formats cover essentially all audio. They divide cleanly into two groups, and the choice within each group is easier than it looks.
Published
The two groups
Keep everything: WAV stores raw uncompressed samples, roughly 10 MB per stereo minute. FLAC compresses those same samples losslessly to about half the size — decode it and you get the original back exactly.
Throw some away: MP3, AAC, Vorbis and Opus discard detail to reach far smaller files. At sensible bitrates the difference is inaudible to most people on most equipment.
Why MP3 refuses to die
Technically, MP3 is the weakest of the lossy formats. AAC beats it at the same bitrate, Opus beats it decisively at low bitrates. Yet MP3 remains the default, for two reasons that have nothing to do with quality.
First, it plays on literally everything — every car stereo, every cheap player, every piece of software written in the last twenty-five years. Second, its patents expired in 2017, so it can be used freely by anyone. Universal compatibility plus zero licensing is a combination none of its technical superiors can match.
When each one is right
- WAV — editing, mastering, feeding hardware or software that demands raw PCM. Not for storage; it is enormous.
- FLAC — archiving anything you care about. Lossless at roughly half the size of WAV, with good tagging.
- MP3 — anything that has to play somewhere you do not control. 192 kbps is a sensible default.
- AAC / M4A — the Apple ecosystem and modern streaming. Better than MP3 at the same bitrate.
- OGG Vorbis — games and open-source projects, where a royalty-free format matters.
- Opus — voice and low-bitrate streaming, where it is comfortably the best available.
Bitrate, briefly
Bitrate is how many bits each second of audio gets. For music, 128 kbps MP3 is audibly compromised on good headphones, 192 kbps is fine for almost everyone, and past 256 kbps you are mostly buying reassurance. For speech, far less is needed — a podcast at 96 kbps mono is perfectly clear.
Variable bitrate spends more on complex passages and less on simple ones, which is strictly better than a fixed rate at the same average. Use it where the option exists.
The one mistake to avoid
Do not convert a lossy file to a lossless one and imagine you have gained anything. An MP3 turned into FLAC is a large file containing exactly the flaws of the MP3. It is a reasonable step if you are about to edit the audio and want no further loss, but it is not restoration, and it is not an archive.
To apply this: convert WAV to MP3 for sharing, convert WAV to FLAC to archive without loss, or compress an audio file to a lower bitrate.