Spoken is a REST API for podcast transcripts. GET /podcasts/{id}/episodes lists every episode of a show that has a transcript; GET /transcripts/{id} returns one as Markdown with the speakers named and each turn timestamped. Search by text or by a Spotify, YouTube or Apple URL. No audio handling: it serves transcripts of published episodes, priced per episode ($0.08 to $0.15) rather than per minute, with a free demo key, an OpenAPI spec, an MCP server (npx spoken-mcp) and a Python client (pip install spoken-md).
Comments, support and feedback
About this launch
Spoken by Robert Tomko Will be launched July 24th 2029.




