formats · 5 min read
MP3 vs WAV vs M4A: which audio format should you use?
WAV stores audio uncompressed, MP3 compresses it with the MPEG-1 Layer III codec, and M4A is normally AAC audio inside an MP4 container. The codec decides quality and file size; the extension only names the wrapper.
Use WAV while recording and editing, MP3 when a file has to play anywhere, and M4A when you want smaller files in the Apple ecosystem. Converting between lossy formats never restores quality.
What each format actually is
A WAV file is a RIFF container, usually holding PCM audio: samples of the waveform at a fixed rate and bit depth, with no perceptual compression. CD-quality stereo means 44,100 samples per second at 16 bits across two channels, which works out to about 1,411 kilobits per second and roughly 10.6 megabytes per minute. A 24-bit recording at 96 kHz needs more than three times the space of the CD-quality version.
MP3 is a lossy perceptual codec. It discards parts of the signal that psychoacoustic models expect listeners not to notice, which is how it reaches bitrates from 32 to 320 kilobits per second. M4A is not a codec but an extension for audio in an MP4 container, usually AAC and occasionally Apple Lossless. AAC is generally more efficient than MP3, so it can sound better at the same bitrate, but it is still lossy.
What a bitrate buys you
Bitrate is how many bits per second the encoder is allowed to spend, and the trade is straightforward: more bits mean more detail retained and a larger file. At 192 kilobits per second an MP3 adds up to roughly 1.4 megabytes per minute, about one-seventh of the uncompressed WAV it came from. Variable bitrate encoding spends more on complex passages and less on quiet ones, which suits music better than a fixed rate.
Speech survives low bitrates well because its useful information is concentrated in a narrow band, which is why voice recordings can be clear at 64 to 96 kilobits per second mono. Detailed music, cymbals and layered mixes need several times that before most listeners stop noticing artefacts. Pushing a lossy encoder beyond its useful range does not add detail back, and transcoding one lossy format to another stacks a second round of loss on the first.
Editing headroom and generation loss
Editing is where uncompressed audio earns its size. Every decode, effect and volume change is arithmetic on samples, and starting from 24-bit PCM gives the processing room to breathe without amplifying noise. A lossy source can still be edited, but the discarded information is gone: converting an MP3 to WAV makes a bigger file, not a better one.
The practical rule is to edit once and encode once. Convert a compressed file to WAV if the editor requires it, do all the cutting and processing, then export the final delivery format in a single pass. Saving an MP3 repeatedly re-encodes it each time, and several generations of that can turn clear speech into a boxy, watery version of itself.
Compatibility on real devices
MP3 has the widest reach of any audio format. Car head units, portable players, old phones, podcast hosts, browsers and operating systems have supported it for decades, which is why it remains the default for anything that must work without asking questions. WAV is widely supported too, especially by editing and DJ software, but its size makes it awkward for streaming, email and upload forms with limits.
Uncompressed is not the only way to keep everything. FLAC compresses PCM losslessly, typically shrinking a WAV by a noticeable margin while decoding back to identical samples, and ALAC does the same inside an M4A container. Both are reasonable archive masters; neither can make a lossy source lossless after the fact.
M4A and AAC are strongest in the Apple ecosystem: iPhones, iPads, macOS, iTunes and Safari handle them natively, and Android, Windows and modern browsers decode AAC through software. Older car stereos and cheap MP3 players are the usual weak point. FLAC is the lossless alternative when both quality and cross-platform support matter, at the cost of much larger files.
Choosing by job
Match the format to the destination rather than picking a favourite. The list below covers the common cases, with bitrates as practical starting points rather than rules.
- Recording and editing: WAV, ideally 24-bit, kept as the working master.
- Podcast or lecture delivery: MP3, mono 64 to 96 kbps for pure speech or stereo 128 to 192 kbps if there is music.
- Music for phones and general listening: MP3 at 192 to 320 kbps, or M4A/AAC around 256 kbps.
- Archiving: FLAC or WAV; never an already-transcoded lossy file.
- Voice memos and Apple recordings going to older hardware: convert M4A to MP3.
- Soundtracks from video: AAC inside MP4 for playback, MP3 when a podcast host or player needs it.