sudo ./install.sh
Copies all scripts and latest sound-gambit binary (if needed) to usr/local/bin
.
Scripts to batch normalize files to integrated or true peak targets
Prerequisites: ffmpeg
, sound-gambit
(https://github.com/x42/sound-gambit)
Fully supported file inputs: wav, aiff (or aif), flac.
Mp3, opus, ogg and wavpack are first converted to wav before normalizing.
Files are analyzed by ffmpeg
with the required gain passed to either ffmpeg
or sound-gambit
.
+/− gain is calculated by the target level minus the analyzed integrated or peak value.
In the case of ebu-norm
, it uses two analysis passes to ensure that the exact integrated value is reached.
Files are written to a sub-folder with suffix added to filename. User-friendly variables near the top of the ebu-norm
script are present to allow for tweaking of the limiter settings.
When using ebu-norm
with large positive gain amounts into the limiter, it should be noted that excessive limiting is no substitute for correctly mastered files. If you find yourself applying more than a couple of dB of peak limiting, perhaps it is a sign to return to the original file in order to use a compressor plugin or to apply manual gain-riding.
ebu-norm [-t target_value] infiles
where -t
allows for an integrated target other than -23 LUFS.
If true peaks would rise above -1 dBTP, the sound-gambit
true-peak limiter is engaged.
tp-norm [-t target_value] infiles
where -t
allows for an true peak target other than -1 dBTP.
ebu-norm AudioFolder/*.wav
will create an ebu-norm
sub-folder and create -23 LUFS integrated WAV files (default).
ebu-norm -t -16 AudioFolder/*
will create an ebu-norm
sub-folder and normalize all supported audio files to -16 LUFS integrated with true-peak limiting as required.
tp-norm AudioFolder/*.wav
will create a tp-norm
sub-folder and -1 dBTP WAV files (default).
tp-norm -t -2 AudioFolder/*.flac
will create a tp-norm
sub-folder and -2 dBTP FLAC files.
ebu-scan [-a] infiles
Script to batch analyze audio files and print true peak and various loudness values to screen and text file. If the -a
flag is added, all files are treated as part of an album and total loudness results are additionally returned. In this instance all files should be the same file format, bit depth, channel count etc.
Prerequisites: ffmpeg
Fully supported file inputs: wav, aiff (or aif), flac, ogg, opus, mp3 and wavpack.
ebu-scan AudioFolder/*.wav
Sample formatted terminal output (and also written to analysis.txt):
File True Peak Integrated Short-term Momentary LRA
(dBTP) (LUFS) (LUFS) (LUFS) (LU)
02 Ubi Caritas.wav -3.2 -21.7 -14.4 -13.1 15.8
07 Bring Him Home.wav -9.0 -25.0 -18.6 -17.5 11.5
08 Steal Away.wav -7.2 -22.4 -16.7 -15.4 12.9
09 Lullaby.wav -2.5 -22.4 -15.0 -11.8 14.7
11 Over the Rainbow.wav -6.8 -21.9 -16.6 -14.8 11.9
Wrapper script to batch produce PNG loudness graphs using adapted PowerShell script.
A huge thanks to user L5730 from https://www.audiosciencereview.com/ for the original PowerShell script.
====Will move to native bash script for v3.0====
Prerequisites: pwsh
(PowerShell), ffmpeg
, gnuplot
Fully supported file inputs: wav, aiff (or aif), flac, ogg, opus, mp3 and wavpack
ebu-scan AudioFolder/Sympathique.flac