Particle launches Radar to make podcast audio searchable for AI
Particle, the AI newsreader startup founded by former Twitter engineers, has launched Radar, a podcast search and intelligence service designed to make spoken audio discoverable by people and AI agents. The company says Radar transcribes more than 130,000 podcasts, including all Apple Top 200 podcasts across 135 verticals, and adds 20,000 episodes to its index every day.
Radar is available through a web interface, but Particle positions its API and Model Context Protocol (MCP) integration as the core product. The system is intended to give businesses and agents programmatic access to information embedded in podcast conversations rather than leaving that material outside text-focused web indexes.
Audio layer for agent workflows
Particle co-founder and CEO Sara Beykpour said many API agents and services crawl the web with a focus on text, while audio remains difficult for them to use unless it has first been transcribed. Radar combines transcription with semantic understanding, allowing users to search for key quotes, highlights and the entities discussed in an episode.
Its transcripts include speaker labels and metadata for people, companies, brands, products and topics. Radar can track entity mentions across podcasts and issue alerts immediately or as daily and weekly digests. Alerts can be delivered by email, Slack or webhook, with filters for conditions such as a specified guest discussing a particular subject or a search limited to top podcasts.
Clips, advertising data and commercial use
The service can return self-contained clips with timestamps, so users can listen to the relevant passage and read the associated transcription. It also tracks topics, when people or organizations are mentioned, listener ratings and reviews, episode advertising, chart rankings, audience-size estimates, sponsorship data, political-bias analysis and brand suitability.
A dedicated advertising search tool identifies episodes in which a company advertises and tracks those appearances over time. Particle says hedge funds have been the highest-volume customers directly integrating Radar’s API, while AI search platforms, data resellers, journalists and researchers are also potential users. Exa is listed among Radar’s partners.
Pricing and next steps
Radar costs $29 per seat each month. Its business plan costs $399 per month for 20 seats, while API access is priced individually based on customer requirements. The product grew from a Particle newsreader feature that surfaced podcast clips next to related stories, before the company shifted toward an API-led podcast intelligence offering.
Particle plans to broaden Radar beyond podcasts to other audio sources, including YouTube videos and news clips. For businesses, the immediate implication is that podcast monitoring can become a structured data workflow, making spoken mentions, advertising activity and expert commentary easier to evaluate alongside text-based research.