About podQuery
We build the semantic layer for spoken media. Podcasts produce more recorded ideas than any other medium — and they're the least retrievable. podQuery turns any podcast archive into something you can search, cite, and reuse: by what was said, who said it, when, and with what evidence.
What we build
A retrieval-grade data layer for audio: speech transcription, speaker attribution, named-entity timelines, summaries and source-grounded fact-check provenance — queryable from a dashboard, a GraphQL API, or any AI client speaking MCP. podQuery is a subscription: every monthly plan includes a quota of ingestion minutes, and additional points can be bought for back catalogues or large corpora. Organisations sponsoring shows for the open Research Lab use the same mechanism or a Custom plan.
The data podQuery computes has a free public face: Research for free on PodBase →
Who it's for
Organisations that treat their archive as an asset, not as old content. Newsrooms with long-form podcasts that need to cite past coverage in minutes, not hours. Publishers and broadcasters with hundreds or thousands of episodes that are de facto unreachable. Agencies managing multiple client podcasts. Universities and research projects with interview archives. And the individual shows serious enough to want their work to outlast the publish day.
What we don't do
We don't host audio, we don't publish full transcripts, and we don't replace your podcast player. We index audio that's already publicly syndicated, surface short framed excerpts under EU copyright law, and link back to your official audio URL. Traffic flows to the producer, not away from them.
Contact
Custom plans, long-term ingestion, sponsoring shows for the Research Lab, partnership inquiries, or general feedback:
contact@podquery.ai