I have made a collection of scripts to help me add chapters to a music podcast I like. They're gross, poorly-written, and generally unsustainable, but they work alright so far.

• chaptagger.py: Version 1, reads a JSON file and writes it into the MP3 file. Usage: python3 chaptagger.py podcast.mp3 data.json. I forget the JSON format, will document later
• chaptagger2.py: Version 2, does the same thing as version 1 but less buggy. It appears to have some debug information in it from when I was testing with Wait Wait... Don't Tell Me!
• chaptagger3.py: Version 3, reads data from an Audition XMP sidecar and writes it to an MP3 file. Usage: python3 chaptagger3.py -m podcast.mp3 -x metadata.xmp (Currently the one I use)
• chaptagger4.py: Version 4, does the same thing as version 3 but with JSON input instead of XMP.
• log2json.py: Convert all of the IRC logs in a directory to the JSON format from v1 and v4 above
• logsub.sh: Read an IRC log and substitute out all of the messages from users who aren't the track announce bot
• mp3chapter-sopel.py: A Sopel bot to make JSON-formatted data for v1 and v4
• timefixer.py: Shift times forward by the exact length of the podcast's intro track for when I didn't get the bot recording fast enough.
• transformer.py: A python script to shift times by some function to fix errors inherent in scraping from IRC. Not very good at its job.
• xmp2lrc.py: Convert XMP sidecars to LRC files, for a friend