Pull the audio out of a video, turn M4A, OGG or WAV into MP3, and cut it on a waveform. Everything runs in your browser, so client recordings never leave your computer.
Output is MP3 or WAV. This page cannot write M4A, AAC, OGG or FLAC, and it cannot fetch a YouTube link. Your file is read from your disk and never uploaded.
Nearly every "extract audio from MP4" result is a site that wants your file on their server first, with a size cap and a queue attached. That is a bad trade for a client interview, a coaching call or an unreleased track. This page uses the audio decoder that is already built into your browser, so the file is read from your disk, decoded in the tab and written back out as a download. Your audio never touches a network.
The same path handles the conversions people actually search for: M4A to MP3, MOV to MP3, WebM to MP3, WAV to MP3 and OGG to MP3. Drop up to 25 files and they come back as a ZIP.
It reads whatever your browser can decode. In practice that is MP4, MOV, M4V, WebM, M4A, AAC, MP3 and WAV everywhere, plus OGG and FLAC on Chrome, Edge and Firefox. Safari refuses OGG and FLAC. AVI, WMV, FLV and most MKV files will not decode in a browser at all, so they fail here the same way they fail in every other browser tool.
It writes MP3 or WAV only. It cannot produce M4A, AAC, OGG, FLAC or WMA, because there is no encoder for those in a browser without a large download. If you came here for MP3 to M4A or WAV to FLAC, this is not the tool. It also cannot take a YouTube or a Vimeo link, only a file already on your device.
Drop a single file and the whole thing is drawn as a waveform. Drag anywhere to sweep out the part you want, then drag either edge to fine tune, or type exact times into the start and end boxes. Press Play to hear only the selection before you commit. The short fade at each cut point is on by default because a hard cut through the middle of a wave makes an audible click.
Handy for cutting an intro off a podcast, pulling one quote out of a client call, or making an Android ringtone. Not an iPhone ringtone: iOS needs an M4R file, which is AAC in an MP4 container, and no browser can write one.
For voice, 128 kbps mono at 44.1 kHz is indistinguishable from the source and about a megabyte a minute. For music, 192 kbps stereo is the sensible default and 320 is worth it only when the source is already lossless. Encoding above the bitrate of your source cannot add quality back, it just makes a bigger file, so the page warns you when you are about to do that. Choose WAV when the audio is going into an editor or a transcriber next, since every conversion to MP3 loses a little.
The MP3 encoder is plain JavaScript running on your own processor at roughly four times real time, so a five minute clip takes about a minute and a half and a one hour lecture takes the best part of fifteen minutes. The progress bar is honest, and you can cancel while it is encoding. Decoding a long file happens inside the browser and cannot be interrupted, so pick the file you want before you start.
The real limit is memory, not speed. Decoding turns a compressed file into raw samples, which for one hour of 48 kHz stereo is around 1.4 GB before anything else happens. The page warns you above 200 MB and lets you decide. Picking Mono halves it, and a lower sample rate cuts it further. If a very long recording refuses to load, split it first or run it at 22.05 kHz mono.
Background hiss or room noise? The free Noise Reducer in the main app cleans a recording and exports WAV. Need the words instead of the sound? The Audio & Video to Text transcriber takes the MP3 you just made. Turning a clip into something postable is the Audiogram Maker, and shrinking the video itself is Compress Video to Target Size.