mp3-split is a command-line utility that splits a long MP3 file into smaller files and then tags them with the track name and track number.

• Python 3.7+
• PyYAML
• python-ffmpeg

ffmpeg is an external dependency and needs to be present on your path.

Usage: mp3-split INPUT_FILE CONFIG_FILE

Example CONFIG_FILE layout:

artist: "A band"
title: "An album"
tracks:
  - name: Track Name 1
    offset: "0:00"
  - name: The Second Track
    offset: "3:26"
  - name: A Third Piece
    offset: "26:01"

Files will be created in [ARTIST]/[ALBUM]/, which will be created if it does not exist.

Setting the LOGLEVEL=DEBUG environment variable will show debug output.

• Converted to use Poetry, making the script installable.
• Included the track number in the filename.
• Removed the need for subprocess shells.
• Removed dependencies on eyeD3 and mp3info, handling these using the Python ffmpeg bindings instead.
• Significant code cleanup including a ~50% length reduction.
• Added doc strings to all functions.
• Added a VSCode devcontainer configuration for easier dependency management.
• Added pre-commit checks.
• Converted the config file to YAML, enabling easier parsing and adding more optional tags, starting with artist and album name.
• Quietened output by default, using the logging module to specify log level.
• Converted to use Click for CLI handling, simplifying argument parsing and file access.
• Removed a lot of unnecessary re-computing of information already in the config file; only the total length needs to be computed.

• Remove the external dependency on the ffmpeg binary, such as by switching to PyAV.
• Add tests.
• Genericise to work on non-MP3 files where possible.

• Original author: Daniel Teichman https://github.com/danielteichman
• Significant rewrite by Chris Cunningham https://github.com/thumperward