Podcast Audio Levels Chart
Most amateur podcast audio problems come down to levels: recording too hot and clipping, recording too quiet and adding noise when boosting later, or publishing at a loudness that sounds wildly different from every other show in the listener's feed. This chart covers the targets at each stage, from recording through the final exported file.
Recording levels
Average speaking level: aim for roughly -18 to -12 dBFS on your recording meter. This leaves plenty of room above for laughter, emphasis, and sudden loud moments without clipping.
Peaks: keep the loudest moments below about -6 dBFS. Digital audio cannot exceed 0 dBFS, and anything that hits it is permanently clipped, producing harsh distortion that no editing can fully remove.
With 24-bit recording, there is little penalty for recording conservatively quiet, since the extra dynamic range means you can raise the level in editing without meaningfully raising the noise floor. Recording slightly low is always recoverable; clipping never is.
Published loudness standards (LUFS)
LUFS (Loudness Units relative to Full Scale) measures perceived loudness across the whole episode, which is what listeners actually experience, rather than momentary peaks.
-16 LUFS integrated: the widely used target for stereo podcast files and the loudness Apple Podcasts recommends. The common default for most shows.
-19 LUFS integrated: the equivalent target for mono podcast files, since a mono file plays back roughly 3 dB louder than the same content in stereo.
-14 LUFS: roughly where Spotify normalizes music playback. Some podcasters master slightly louder for Spotify-first shows, though -16 remains the safe general standard.
True peak ceiling
Set a true peak limit of -1 dBTP on the final export. Standard peak meters can miss inter-sample peaks that occur when the file is converted to MP3 or AAC, and those peaks can clip after encoding even when the uncompressed file looked safe.
A limiter set to -1 dBTP as the last processing step handles this automatically, and is one of the easiest and most overlooked fixes for distortion that only appears once an episode is published.
Recording format settings
Sample rate: 44.1 kHz or 48 kHz. 48 kHz is the video standard and the better choice if the show is ever published to YouTube or cut with video. Higher sample rates add file size without any audible benefit for spoken word.
Bit depth: record at 24-bit for the headroom advantage described above, even though the final published file will be compressed.
Final export: MP3 at 128 kbps is sufficient for mono speech and 192 kbps for stereo or shows with music. Record in WAV or another lossless format, and only compress once at the very end, since re-encoding an already compressed file stacks quality loss with every pass.