How to choose, trim and level podcast intro music

To add intro music to a podcast, choose a short track, trim it to length, fade it out, and level the finished episode to about -16 LUFS with a true peak of -1 dBTP or lower. The ffmpeg commands below do the trimming, fading and measuring.

Last updated:

Example tracks

These are real catalog tracks. The credit line under each row is the one Uncommon Ear generates.

  • Monplaisir Mid-tempo. Suggested for loop.

    0:19

  • Abstraction Fast. Suggested for intro / outro.

    0:20

  • Emma_MA Mid-tempo ambient. Suggested for loop, intro / outro.

    0:27

  • n2p5 Credit required Fast jazz. Suggested for intro / outro.

    0:13
    Credit to include

    “Fanfarejazz1” by n2p5, via Freesound, licensed CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ https://freesound.org/s/49478/

  • Kenney Suggested for stinger.

    0.3 s

Steps

  1. Choose a track

    For an intro, look for a track of 10 to 30 seconds with a clear ending, or a loop you can cut. Listen to it under your voice before you commit, because a busy melody competes with speech. Check the license badge. A CC0 track needs no credit, and a CC BY track needs its credit line in your show notes.

  2. Trim and fade

    Cut the track to the length you want and add a short fade-in and a longer fade-out so it does not end abruptly. This command starts 5 seconds in, keeps 15 seconds, fades in over half a second and fades out over the last 3 seconds. The fade-out start time is counted from the start of the trimmed clip.

    Trim and fade 
    ffmpeg -ss 5 -t 15 -i intro.wav -af "afade=t=in:st=0:d=0.5,afade=t=out:st=12:d=3" intro-15s.wav

  3. Measure the loudness

    Run the loudnorm filter in measure mode to read the integrated loudness in LUFS and the true peak. The Input Integrated and Input True Peak lines describe your file as it is now.

    Measure the loudness 
    ffmpeg -hide_banner -nostats -i intro-15s.wav -af loudnorm=I=-16:TP=-1:LRA=11:print_format=summary -f null -

  4. Match the music to your target

    Apple Podcasts recommends an overall loudness around -16 LKFS, with a tolerance of 1 dB, and a true peak that does not exceed -1 dBFS, measured with ITU-R BS.1770-5. LKFS and LUFS are the same unit. This command brings the music to that target. Set the sample rate yourself, because loudnorm outputs 192 kHz audio if you do not.

    Match the music to your target 
    ffmpeg -i intro-15s.wav -af loudnorm=I=-16:TP=-1:LRA=11 -ar 44100 intro-level.wav

  5. Mix, then measure the whole episode

    Place the intro and outro in your editor and lower the music under any speech by ear. Then measure the finished episode with the same loudnorm command, and adjust the mix until the overall figure is near -16 LUFS. If your show is mono, some producers use -19 LUFS, because many devices play a mono file on two speakers without the 3 dB reduction that keeps it equally loud.

  6. Export for your host

    Apple says the loudness steps must happen before encoding. Apple Podcasts accepts MP3 or AAC for RSS feeds, and recommends 128 to 256 kbps for stereo MP3. This command makes a 128 kbps MP3.

    Export for your host 
    ffmpeg -i episode-mix.wav -codec:a libmp3lame -b:a 128k episode.mp3

  7. Credit the music

    If the track is CC BY, add its credit line to your episode notes and say if you trimmed it, for example: Shortened to 15 seconds.

Common mistakes

  • Leveling the intro alone and never measuring the finished episode. Loudness targets apply to the overall mix.
  • Normalizing twice. Each pass can change the peak and the loudness range, so measure once at the end.
  • Running loudnorm without a sample rate. The output is 192 kHz, far above the 44.1 or 48 kHz that Apple Podcasts lists for MP3 and AAC.
  • Cutting the intro mid-phrase with no fade. Use a fade-out of a second or more.
  • Using a CC BY-NC track in a podcast with ads or sponsors. NC licenses do not allow commercial use.
  • Encoding to MP3 before measuring. Do the loudness work on the WAV, then encode.