Exa Adds 3.2M Episodes to Search API
Exa now indexes 3.2 million podcast episodes across 119,000 shows, with full speaker-labeled transcripts and timestamps. The dataset is searchable through Exa’s existing API, enabling agents to discover, summarize, and recommend episodes based on what was actually said.
Exa turns podcast discovery from a browsing task into a retrieval problem that agents can handle exceptionally well. The real differentiator is transcript-level search, though coverage, licensing, and transcription accuracy remain important caveats.
- –Full transcripts expose topics, guests, and exact passages that episode titles and descriptions miss.
- –Developers can use the existing `/search` endpoint and API key, with no separate podcast integration.
- –Exa reports 98.4% freshness and 85.0% answer correctness in its own seven-API benchmark.
- –The feature supports personalized recommendations, podcast RAG, executive monitoring, recruiting, and market research.
- –Speaker labels and timestamps make grounded answers and quote extraction substantially more useful.
DISCOVERED
1h ago
2026-11-02
PUBLISHED
1h ago
2026-11-02
RELEVANCE
AUTHOR
rileybrown