ffmpeg-normalize 0.5

Normalize audio via ffmpeg

ffmpeg-normalize

ffmpeg script for normalizing audio.

This program normalizes media files to a certain dB level. The default is an RMS-based normalization where the mean is lifted or attenuated. Peak normalization is possible with the -m option.

It takes any audio or video file as input, and writes the audio part as output WAV file. The normalized audio can also be merged with the original.

Requirements

• Python

• ffmpeg from http://ffmpeg.org/ installed in your $PATH (3.x or above recommended)

Installation

pip install ffmpeg-normalize

Usage

ffmpeg-normalize [options] <input-file> ...

Options

Normalization:

• -l, --level <level>: dB level to normalize to [default: -26]

• -m, --max: Normalize to the maximum (peak) volume instead of RMS

• -b, --ebu: Normalize according to EBU R128 (ffmpeg loudnorm filter)

• -t, --threshold <threshold>: dB threshold below which the audio will be not adjusted [default: 0.5]

Encoding / Format:

• -a, --acodec <acodec>: Set audio codec for ffmpeg (see ffmpeg -encoders) to use for output (will be chosen based on format, default pcm_s16le for WAV)

• -r, --format <format>: Set format / file extension for ffmpeg (see ffmpeg -formats) to use for output file [default: wav]

• -e, --extra-options <extra-options> Set extra options passed to ffmpeg (e.g. -b:a 192k to set audio bitrate)

File Handling:

• -f, --force: Force overwriting existing files

• -p, --prefix <prefix>: Prefix for normalized files or output folder name [default: normalized]

• -x, --no-prefix: Write output file without prefix (cannot be used when --dir is used)

• -o, --dir: Create an output folder under the input file’s directory with the prefix instead of prefixing the file (does not work if --no-prefix is chosen)

• -u, --merge: Take original file’s streams and merge the normalized audio. Note: This will not overwrite the input file, but output to normalized-<input>.

General:

• -v, --verbose: Enable verbose output

• -n, --dry-run: Show what would be done, do not convert

• -d, --debug: Show debug output

Examples

Normalize a file and write to normalized-file.wav:

ffmpeg-normalize -v file.mp3
ffmpeg-normalize --verbose *.avi

Normalize a number of AVI files and write to normalized-<file>.wav:

ffmpeg-normalize -v *.avi
ffmpeg-normalize --verbose *.avi

Normalize a number of MP4 files to -5 dB peak volume and merge the audio stream back into the MP4 files, in a new directory called normalized:

ffmpeg-normalize -vuofm -l -5 *.mp4
ffmpeg-normalize --verbose --merge --dir --force --max --level -5 *.mp4

Normalize a number of MKV files and merge the audio back in using the libfdk_aac encoder with 192 kBit/s CBR:

ffmpeg-normalize -vu -a libfdk_aac -e "-b:a 192k" *.mkv
ffmpeg-normalize --verbose --merge --acodec libfdk_aac --extra-options "-b:a 192k" *.mkv

License

The MIT License (MIT)

Copyright (c) 2015-2017 Werner Robitza

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

History

0.5 (2017-04-02)

• Add new EBU R128 normalization filter

• Fix issue with output file extension not being WAV by default

• Fix issue #24 where setup.py fails on Windows / Python 3.6

0.4.3 (2017-02-27)

• Fix option -np, should be -x short

• Abort when input and output file are the same (ffmpeg can’t overwrite it)

0.4.2 (2017-02-27)

• Map metadata from input to output when merging

• Clarify use of merge option

0.4.1 (2017-02-13)

• Fix #13

0.4 (2017-01-24)

• Cleanup in code, make it class-based

• Drop avconv support, it was never good anyway

• Add support for specifying codec for non-merging operations

• Add support for specifying output format

• README improvements

0.3 (2017-01-19)

• Add option to remove prefix

0.2.4 (2016-10-27)

• Fixed issue where multiple spaces were collapsed into one

0.2.3 (2016-02-12)

• Fixed issue where ffmpeg could not be found if path included spaces

0.2.2 (2016-02-09)

• Change default level back to -26

0.2.1 (2016-02-08)

• Documentation fixes

0.2.0 (2016-02-08)

• Support multiple input files

• Allow merging with input file instead of creating separate WAV

• Write to directory instead of using prefix

• Set the audio codec when merging

• Set additional encoder or ffmpeg options

Note: avconv support is very limited, use the real ffmpeg from http://ffmpeg.org/ instead.

0.1.3 (2015-12-15)

• Bugfix for detecting ffmpeg or avconv on Windows (as .exe)

• Add version to Usage message

• Update year

0.1.2 (2015-11-13)

• Bugfix for missing ffmpeg or avconv

0.1.0 (2015-08-01)

• First release, changing name to ffmpeg-normalize