How to save an Audacity file as an MP3

File and Save Project produces an .aup3 file. That is Audacity's working format: your tracks, your edits, your labels, everything still separate and still editable. Only Audacity can open it.

File and Export produces audio: MP3, WAV, and the rest. That is the file you send to someone, upload to a host, or play on a phone. Everything is mixed down and nothing can be edited apart any more.

So if you have emailed someone a file they cannot open, you sent the project. Go back and use Export instead.

Do both. Save the project so you can come back and change something, and export the MP3 to publish. They serve different purposes and neither replaces the other.

File, Export, Export as MP3. Choose where it goes and name it. In current versions this appears as Export Audio with a format dropdown, so pick MP3 there.

Audacity no longer needs the LAME encoder installed separately. If you are following an older tutorial that tells you to download it, ignore that step, because MP3 export has been built in for several versions now.

If nothing is selected, Audacity exports the whole project. If you have a selection, look for the option to export only the selected audio, which is easy to leave ticked by accident from a previous export.

Bit Rate Mode: Constant is predictable and universally compatible. Variable gives slightly better quality for the same size but some older podcast tools handle it poorly. For podcasts, use Constant.

Quality, meaning bitrate: 128 kbps stereo is the standard for spoken audio and is what most podcast hosts expect. 96 kbps mono is perfectly good for a single voice and roughly halves the file size. 192 kbps and above is for music and is wasted on speech.

Channel Mode: for one person talking into one microphone, Mono is the honest choice. Stereo doubles the file size to store two identical channels. Use Joint Stereo if there is music or if two people are genuinely placed left and right.

Fill in the metadata dialogue that appears next: title, artist, year. It is what shows in a player, and it takes ten seconds now versus editing tags later.

Export captures whatever level your project is at. If the mix is quiet, the MP3 is quiet, and fixing it afterwards means exporting again.

Before exporting, select everything and run Effect, Volume and Compression, Loudness Normalization, targeting about minus 16 LUFS for stereo or minus 19 for mono. That is what podcast platforms expect, and delivering at it means nothing gets adjusted on the way in.

Check the very end of the project too. A few seconds of silence after the last word is fine; thirty seconds of dead air is a surprisingly common thing to ship.

An export that produces silence usually means a track was muted, or a solo button was left on somewhere else in the project. Check every track's mute and solo state before assuming the export failed.

An MP3 much shorter than expected means you exported a selection rather than the project. Click somewhere in the waveform to clear the selection, or use Select and All, then export again.

Distortion in the MP3 that was not there in Audacity means the mix was clipping. Turn on View and Show Clipping, look for red, lower the track gains, normalise and export again.

Always play the exported file before sending it. Open the MP3 itself, not the project, and listen to the start and the end at minimum.