MP3 is the format the internet forgot to retire. It is larger than the newest codecs, smaller than uncompressed audio, and it plays absolutely everywhere, which makes it the safe answer to most “just send me the sound” requests.
What the conversion does
A video file contains a video stream and an audio stream in one container. This tool opens the container, takes the audio stream, and re-encodes it as an MP3. The result is a standalone audio file that no longer needs the picture, which is what you want when the visual half was never the point.
Because MP3 compresses, the file comes out dramatically smaller than the same audio stored uncompressed. That reduction is lossy, meaning some detail is genuinely discarded rather than hidden. At a sensible bitrate the loss on speech and music is very difficult to hear, which is why MP3 became the default rather than an accident of history.
Choosing between this and a WAV extraction
If you plan to edit the audio, work with the uncompressed version instead. Noise reduction, levelling and trimming all benefit from starting with everything intact, and re-encoding an already compressed file twice is where quality really starts to fall away. Extract first, edit, and convert to MP3 as the last step.
If you simply need to send or upload the sound, going straight to MP3 is the efficient route. You avoid producing a large intermediate file and get something that plays immediately.
Nothing is uploaded
The work happens in your browser, so the file you drop in is read locally and never sent to a server. That makes it reasonable for recordings you would rather not hand over: an interview that is not ready for release, a rehearsal, something recorded for one person that happens to be in a video file.
After the MP3 exists
Having the audio in one small file makes a few things easier. It can be transcribed to text, cut down to the part that matters, or sharpened up if it was recorded in a quiet room and has become a little dull in the round trip. Each of those works better on a clean standalone file than on a video.
Picking a bitrate
MP3 encoders usually ask for a bitrate, and the number matters more than most people expect. Around 128 kbps is perfectly fine for speech and adequate for music, 192 kbps is comfortably transparent for the majority of listeners, and 320 kbps is where returns mostly stop. Dropping below roughly 96 kbps becomes audible on sustained sounds such as cymbals or a held vocal note. File size scales in proportion, so a three-minute recording moves from about 2.8 MB at 128 kbps to about 7 MB at 320 kbps. For a clip you are emailing or sending in a message, the lower end of that range is the sensible choice.
It is worth keeping the original video as well. If the sound turns out to be the wrong take for the job, or the conversion was made from a clip you were not happy with in the first place, having the source to hand means the mistake costs a few seconds rather than the whole file.