Particle launches Radar to index and search podcast audio for AI agents
Particle, an AI newsreader startup, introduced Radar, a podcast search engine that transcribes and analyzes over 130,000 podcasts, making them searchable and accessible via API. The service is attracting interest from hedge funds and AI search platforms. It provides speaker labels, metadata, and entity tracking.
Radar processes over 130,000 podcasts, adding roughly 20,000 episodes daily, and covers Apple's Top 200 podcasts across 135 verticals. The service provides speaker labels, entity tracking, and pre-selected notable clips with timestamps, allowing users to listen or read specific comments without consuming full episodes.
Beyond search, Radar offers customizable alerts delivered via email, Slack, or webhook, with filters for specific guests or topics. A dedicated ads search engine tracks company advertisements across episodes, while additional tools analyze political bias, audience size, sponsorship data, and brand suitability. Pricing starts at $29 monthly, with the API and MCP serving as the primary commercial product.
Radar could significantly expand how businesses and researchers access spoken content, making podcasts as searchable as written text. Hedge funds and AI platforms may gain competitive advantages through data previously invisible to automated systems, while journalists could locate sources faster. However, increased monitoring of spoken conversations may raise privacy considerations, and smaller podcasters could face pressure as their content becomes commodified for commercial intelligence purposes.