What format should a podcast be?

Record to WAV if you can. It is uncompressed, so nothing is lost before you have even started editing, and every editor opens it without thinking. 48 kHz at 24-bit is a sensible default; 44.1 kHz at 16-bit is fine too and smaller.

If storage is tight, FLAC is lossless as well and about half the size of WAV. FLAC vs WAV covers the trade.

What matters is that you never edit from a lossy file. Every export from MP3 back to MP3 throws away another layer of detail, and speech shows it quickly.

MP3 remains the safe delivery format for podcasts: every app, every car stereo and every smart speaker plays it, and chapter and tag support is well understood.

For a voice-only show, 96 to 128 kbps mono is plenty. Mono is not a compromise for speech — one person speaking has no stereo information to preserve, and mono halves the bitrate needed for the same quality. Stereo to Mono converts a recording that was captured in stereo unnecessarily.

If the show has music beds or two microphones panned apart, go 128 to 192 kbps stereo. Beyond that, listeners on phone speakers and car stereos hear nothing extra and pay for it in download size.

Audio Converter handles the export, in your browser, so unreleased episodes do not get uploaded anywhere.

Loudness. Normalize to −16 LUFS stereo, −19 mono, with true peak at −1 dBTP. This is what stops listeners reaching for the volume between your show and the next one. Normalize Audio measures and applies it.

Silence. Long gaps at the start, end and between takes are dead weight in both attention and file size. Remove Silence trims them, and Trim Audio cuts the top and tail without re-encoding.

One export. Do the loudness and the trimming on the lossless file, then encode to MP3 once at the end.