m4b-merge

---

A CLI tool which outputs consistently sorted, tagged, single m4b files regardless of the input.

📝 Table of Contents

• About
• Getting Started
• Usage
• Built Using
• Contributing
• Authors
• Acknowledgments

🧐 About

m4b-merge was originally part of Bragi Books, but was split apart to allow savvy users to automate its usage in more advanced ways. Some of the things m4b-merge offers are:

• Accepts single and multiple mp3, m4a and m4b files.
• mp3s are converted to m4b. m4a/m4b files are edited/merged without conversion.
• Matches existing bitrate and samplerate for target file conversions.
• Final files moved to /output/Author/Book/Book: Subtitle.m4b format.
• Moves finished files into done folder in input directory.

Metadata provided by audnexus:

• Title, authors, narrators, description, series, genres, release year - written as tags.
• Chapter times/titles (only when input is m4b or a single mp3) - written as tags and chapters.txt.
• High resolution (2000x2000 or greater) cover art - embedded into output file.

🏁 Getting Started

Prerequisites

You can either install this project via pip directly or run it prepackaged in Docker:

• If installing directly on your system, you'll need to install m4b-tool and it's dependants from the project's readme
• If using Docker, all prerequisites are included in the image.

Installing

For a pip installation

Simply run

pip install m4b-merge

For a Docker installation

You'll need to specify input/output volumes in the run command for easy use later:

docker run --name=merge -v /path/to/input:/input -v /path/to/output:/output ghcr.io/djdembeck/m4b-merge:main

You may also specify the user and group to run as with env variables:

-e UID=99 -e GID=100

🔧 Running the tests

• Run pip install pytest
• To run all tests, run pytest from inside this project directory.
• To run a single test, run pytest tests/test_NAME.py

🎈 Usage

Workflow

The process is simple

1. Pass the file as input via -i FILE.ext or folder -i DIR/
2. Enter the ASIN (found on audible.com) when prompted.
3. Depending on necessary conversions, the process will take between 5 seconds and 5-10 minutes.

CLI usage

usage: m4b-merge [-h] [--api_url API_URL] [--completed_directory COMPLETED_DIRECTORY] -i INPUTS [INPUTS ...] [--log_level LOG_LEVEL]
                 [--num_cpus NUM_CPUS] [-o OUTPUT]

m4bmerge cli

optional arguments:
  -h, --help            show this help message and exit
  --api_url API_URL     Audnexus mirror to use
  --completed_directory COMPLETED_DIRECTORY
                        Directory path to move original input files to
  -i INPUTS [INPUTS ...], --inputs INPUTS [INPUTS ...]
                        Input paths to process
  --log_level LOG_LEVEL
                        Set logging level
  --num_cpus NUM_CPUS   Number of CPUs to use
  -o OUTPUT, --output OUTPUT
                        Output directory
  -p PATH_FORMAT, --path_format PATH_FORMAT
                        Structure of output path/naming.Supported terms: author, narrator, series_name, series_position, subtitle, title, year
  --chapters-offset     For files not from Audible, adjust chapters to offset according to brandIntroDurationMs

When installed via pip, you can run inputs like so

m4b-merge -i /path/to/file.mp3

Or for multiple inputs

m4b-merge -i /path/to/file.mp3 /dir/ /path/to/other/file

On Docker, you can run inputs like so

docker run -it merge m4b-merge -i /input/file.mp3

For a folder of multiple audio files, simply pass the folder itself as an input, such as -i /input/dir

⛏️ Built Using

• audnexus - API backend for metadata
• m4b-tool - File merging and tagging

✍️ Authors

• @djdembeck - Idea & Initial work

🎉 Acknowledgements

• sandreas for creating and maintaining m4b-tool