Podcast loudness: what LUFS and true peak mean for listeners
Imagine listening to an interview on your phone. The host sounds fine, but you turn the volume up when the guest begins. Then the music returns and you reach for the controls again. A meter may show an acceptable average for the episode while the listening experience still feels uneven.
Good podcast loudness is not about forcing every moment to sound identical. It is about making the conversation understandable and reasonably consistent. LUFS and true-peak measurements help, but they cannot tell you whether a quiet answer disappears under the music.
What do LUFS and true peak tell you?
LUFS is a unit used to describe perceived loudness. Integrated loudness summarises the programme as a whole; shorter-term measurements help inspect individual passages. An average, however, cannot tell you that every sentence is equally easy to hear.
True peak estimates the highest level a signal can reach between samples. It provides information that is not always obvious from a waveform display alone. Speech balance, room reflections, microphone distance and music placement matter too.
Consider an interview in which the host sits close to the microphone while the guest sits farther away. Raising the whole file makes both louder without fixing the difference between them. Balancing the guest's track or the affected passages first is usually a more useful starting point; measure the complete programme after the adjustment.
Which loudness target should you choose?
Start with your destination. Apple Podcasts' audio guidance recommends an overall loudness of −16 LKFS, within ±1 dB, and a true peak no higher than −1 dB FS. That is guidance for Apple's workflow, not a universal requirement for every form of audio.
The sources use different abbreviations, but ITU’s terminology identifies LUFS and LKFS as equivalent units. The label alone does not mean the target is different; it still matters whether a reading describes the whole programme or a short passage.
For comparison, EBU R 128 describes a −23 LUFS programme target in a broadcast context. Different numbers do not automatically mean one is wrong: production and distribution contexts differ. Consult the current instructions for the broadcaster or distributor receiving your file.
Listen while exploring this story
Tune into campus radio with non-stop independent music and culture.
RadioTEDU offers a useful example of why broadcast references and podcast destinations should be considered separately. Its Technologies page says the station uses ITU-R BS.1770-5 for loudness and true-peak measurement alongside the EBU R128 broadcast reference. That is the organisation’s description of its broadcast practice, not a measurement of an individual podcast episode or a target for every file sent to Apple. Check the receiving platform’s instructions for the file you are preparing.
Listen to the mix before chasing the number
Once the edit is complete, listen through the episode, paying attention to quiet sentences and music transitions. If an introduction masks the host's first words, lowering the music may solve more than raising the entire file. Large differences between speakers deserve attention before final measurement.
Dynamics processing can help manage changes in level, but excessive compression may make breaths conspicuous or flatten the delivery. Make modest adjustments and compare them with the earlier version. A meter is most helpful when it sits alongside listening rather than replacing it.
Suppose a master measures close to Apple's recommendation, yet a guest is hard to follow on a bus. The measurement does not make the problem disappear. Adjust the dialogue-to-music balance and measure the complete episode again.
Check the file people will actually hear
Apple recommends its loudness and peak processing before encoding. After checking the production version, listen to the encoded delivery file as well, including the passages most likely to cause trouble. Headphones and an ordinary speaker can reveal different issues.
On the encoded copy, check the loudest music transition and the guest’s quietest answer, not just the opening minute. If you have a true-peak meter, measure the delivery copy again: lossy encoding can change the peak level relative to the master. Apple’s pre-encoding recommendation does not prevent a final check of the file being delivered. If a measurement reveals a problem, return to the source, adjust it and listen to the new copy.
A good result is more than a target reading: the sound should let the audience stay with the conversation. Measurement and attentive listening work best together.
