chaptermill: a podcast's back catalog to show notes for 3¢ an audio hour, on an old PC I kept crashing on purpose

Description and use

chaptermill converts a podcast archive into transcripts, chapters, verified pull quotes, and markdown show notes using open‑weight models and Temporal for durability, reducing SaaS costs.

Problem

Most podcasts with long archives lack transcripts, chapters, and readable show notes, making it hard for new listeners to discover content and for producers to deliver standard deliverables without expensive SaaS.

Outcome

chaptermill enabled a 2015‑era desktop to process 4.3 hours of podcast audio in 44 minutes, producing all required deliverables at a fraction of SaaS cost, with Temporal ensuring resilience to crashes and API failures.

Workflow

• Point chaptermill at a podcast RSS feed
• CatalogWorkflow reads feed and starts EpisodeWorkflow for each episode
• EpisodeWorkflow downloads audio and splits into 5‑minute chunks
• Each chunk is transcribed locally with faster‑whisper
• Transcripts are merged
• Agent loop runs with outline and tools (read_transcript, search_transcript, submit_notes)
• Each model and tool call is a Temporal activity with retry and timeout policies
• Agent submits notes, which are validated and accepted or rejected
• Files (transcript, chapters, quotes, show notes) are written and made downloadable

Evidence

chaptermill takes a podcast's RSS feed and, on a machine you already own, turns the back catalog into the files a producer actually delivers: a timestamped transcript (SRT, VTT, JSON), chapters as a Podcasting 2.0 chapters.json and as YouTube chapter lines, 2–4 pull quotes, each checked word for word against the transcript before it is accepted, Markdown show notes with a summary, topics and a credit line.

Source