Audio Converter

An audio converter reads a sound file in one format and writes it out in another. The common reason is that something downstream refuses the format you have. M4A is the usual culprit: it is what an iPhone voice memo, a Mac recording and most Apple downloads produce, and plenty of software, car stereos and upload forms still only take MP3.

Worth knowing before you convert: M4A and MP3 are both lossy. M4A holds AAC, which is a more modern codec and generally sounds better at the same bitrate, so an M4A is not a worse file than an MP3 and converting it is not an upgrade. It is a compatibility step, and it costs something. The audio has to be decoded from AAC, which has already had detail removed, then compressed again by a different algorithm that cannot tell which of what it hears was in the original recording and which the first encoder left behind. That is generation loss, and it shows up first on cymbals, applause and sibilance.

You cannot avoid that cost when both formats are lossy, but you can avoid making it worse. Do not drop the bitrate below what the source was: re-encoding a 128 kbps M4A at 320 kbps gains nothing but file size, while re-encoding it at 96 kbps throws away a second helping of detail. Converting once is also better than converting repeatedly, so keep the original rather than the converted copy as your master.

WAV and FLAC are the lossless targets and behave differently. Neither adds further loss, though neither can restore what a lossy source already discarded. WAV is uncompressed PCM, which is what audio editors, transcription services and speech models generally want as input, and it runs to about 10 MB a minute in stereo. FLAC compresses losslessly to roughly 60% of that, which makes it the sensible archive format when the source was lossless to begin with.

All of it happens in the browser through ffmpeg compiled to WebAssembly. The file is read from your disk into the page, converted in the tab, and handed back as a download. Nothing is uploaded and no copy sits on a server afterwards. For a voice memo, an interview or a recorded call, that is the whole difference between a private conversion and sending your audio to a stranger.