MP3, WAV, FLAC, AAC and OGG: Which Audio Format to Use for What
Six formats, three categories, one decision. What each is genuinely for, which conversions are safe, and the one mistake that quietly ruins audio quality.
Updated August 11, 20265 min read
There are only three categories of audio format and once you see which is which, the choice takes seconds. Uncompressed stores every sample exactly and is enormous. Lossless compressed stores every sample exactly and is about half the size. Lossy throws away what you are unlikely to hear and is a tenth of the size or less. Everything else is detail — and one rule about the order you convert in, which is where quality quietly disappears.
The formats, sorted
| Format | Type | Size for 3 min | Best for |
|---|---|---|---|
| WAV | Uncompressed | ~30 MB | Editing, mastering, archival masters |
| FLAC | Lossless compressed | ~15 MB | Archiving, audiophile libraries |
| MP3 | Lossy | ~4 MB at 192k | Universal compatibility |
| AAC / M4A | Lossy | ~3 MB at 160k | Apple devices, video audio, streaming |
| OGG | Lossy | ~3 MB at 160k | Open-source, games, web |
Two things people commonly get wrong about this table. FLAC is not higher quality than WAV — both are mathematically exact, and FLAC is simply the same data packed more efficiently. And AAC is genuinely better than MP3 at the same bitrate, particularly below 128kbps, because it is a newer design (1997 vs 1993), built as MP3's successor.
The rule that matters most
This is the single most common way audio quality gets destroyed, because each individual step seems harmless. A track ripped to MP3, converted to M4A for a phone, converted back to MP3 for a car stereo, then compressed again to email has been through four lossy encoders. It will sound thin and smeared, and no amount of processing brings it back.
The defence is simple: keep a lossless master. Archive in FLAC, and generate every lossy version from that master rather than from another lossy file. Converting FLAC to MP3 and FLAC to AAC gives you two good files; converting MP3 to AAC gives you one bad one.
Which conversions are safe
| From → To | Verdict | Why |
|---|---|---|
| WAV → FLAC | Perfect | Lossless, about half the size |
| FLAC → WAV | Perfect | Exact reconstruction |
| WAV or FLAC → MP3/AAC | Fine | One lossy pass from a clean source |
| MP3 → WAV | Pointless | Large file, no quality recovered |
| MP3 → FLAC | Pointless | Losslessly preserves the damage |
| MP3 → AAC | Avoid | Two lossy passes stacked |
| AAC → MP3 | Avoid unless required | Same problem, sometimes unavoidable |
The two "pointless" rows catch a lot of people, because they look like upgrades. Converting an MP3 to WAV does not restore anything — it stores the already-degraded audio uncompressed, giving you a file ten times larger that sounds identical. The information left when the MP3 was made.
Bitrate: how much to spend
- 320 kbps — transparent for essentially everyone. Use when storage is not a concern (a Pro setting in our converter; Free goes up to 192).
- 192 kbps — the sensible default for music. Very few people distinguish it from the source on ordinary equipment.
- 128 kbps — acceptable for casual listening; artefacts appear on cymbals and applause.
- 96 kbps and below — fine for speech, poor for music.
- 64 kbps mono — the right choice for spoken-word podcasts and lecture recordings, where it saves enormous space at no meaningful cost.
Speech and music have genuinely different needs and treating them the same wastes space or quality. A two-hour lecture at 192kbps stereo is 170MB for no benefit — the same recording at 64kbps mono is 55MB and sounds the same, because there is nothing stereo about one person talking. (Our free converter takes files up to 10 minutes and 50 MB, so a lecture that long needs Pro, which removes both caps.)
MP3, WAV, OGG, M4A, FLAC and AAC, converted on your own device, not our servers.
Convert audioSample rate and bit depth
Two more numbers that appear in conversion dialogs and are usually best left alone. Sample rate is how many times per second the waveform was measured — 44.1kHz for CD, 48kHz for video. Bit depth is how precisely each measurement is stored — 16-bit for distribution, 24-bit while recording and editing.
Raising either after the fact adds nothing: converting 44.1kHz to 96kHz interpolates between existing samples and creates no new information, exactly as enlarging a photo does not add detail. The one case where it genuinely matters is matching your target — audio destined for video should be 48kHz, and mismatches there are a classic cause of slow drift out of sync over a long recording. The resample tool handles that conversion cleanly.
Picking in ten seconds
- 1Editing or mastering? WAV. You want no compression in the working file.
- 2Archiving your own masters? FLAC. Exact, and half the size of WAV.
- 3Sending to someone, destination unknown? MP3 at 192kbps. It plays on everything.
- 4Apple ecosystem, or audio for video? M4A/AAC. Better quality per byte and native everywhere it matters.
- 5Speech recording? MP3 or M4A at 64kbps mono. Do not pay for stereo music quality on one voice.
And whichever you choose, keep the original. Every conversion is a one-way door if it is lossy, and the file you wish you had kept is always the one you deleted to save 40MB.
Frequently asked questions
Which audio format has the best quality?
WAV and FLAC are both mathematically exact, so neither loses anything — FLAC is simply the same data compressed to about half the size. Among lossy formats AAC beats MP3 at the same bitrate, especially below 128kbps, because it is a newer design (1997 against MP3's 1993), built as its successor.
Does converting MP3 to WAV improve quality?
No. The detail was discarded when the MP3 was created and cannot be recovered. Converting to WAV stores that already-degraded audio uncompressed, producing a file roughly ten times larger that sounds exactly the same.
Is it bad to convert MP3 to AAC?
Yes, where it is avoidable. Each lossy encoder discards different things, so stacking two passes compounds the damage. Always generate lossy files from a lossless master rather than from another lossy file — keep a FLAC or WAV copy for exactly this reason.
What bitrate should I use for MP3?
192kbps is the sensible default for music and is transparent to most listeners on ordinary equipment; 320kbps if storage is not a concern (a Pro setting in our converter; Free goes up to 192). For speech, 64kbps mono is plenty and saves a great deal of space on long recordings.