Podcast Transcription Platform: Self-Hosted Whisper

Paste a podcast link from Apple Podcasts, Spotify or RSS and get back a transcript, summary, chapters and FAQs. Transcripts go into a public, searchable library, and a paid API gives developers the same engine.

I took over the backend in September 2025 and have written about 576 of the 593 backend commits since then, including the move to self-hosted transcription.

Self-hosted transcription: replaced a paid speech-to-text API with whisper.cpp running on an in-house GPU server.

Job queue across two servers, with paid, free and crawler tiers and crash recovery.

Long-audio processing in chunks, with corrected timestamps and real progress reporting.

AI summaries from a local model and a hosted model, under a daily spending cap.

Back-catalogue crawler that transcribes older episodes while the GPU is idle.

Episode features: speaker labels, search, chat with an episode, and export to PDF, SRT, VTT, Notion and Obsidian.

Public API with credit plans, plus a second brand running on the same backend.

A clean-up step that blocked the server: removing Whisper's repeated lines had quadratic cost and froze the server for minutes on long episodes. I rewrote it to run in bounded time.

Billing you can trust: API refunds are claimed exactly once, and an append-only ledger records every credit change.

52,700+ public transcripts.

41,000+ hours of audio transcribed.

73 languages.

3,200+ backend tests.

Live: https://podcasttranscript.ai