TechCrunch Startup News Particle’s AI news app listens to podcasts for interesting clips so you you don’t have to
Feb 24, 2026
An AI app that pulls short podcast clips into news feeds is explained in detail. The conversation covers how vector embeddings and AI match audio to stories. Listeners hear about transcription, proprietary clipping tech, and newsrooms using AI to monitor podcast commentary. The service’s subscription perks, Android launch, improved browsing, and global audience are also highlighted.
AI Snips
Chapters
Transcript
Episode notes
Podcast Clips Pull Quotes Directly Into News Feed
- Particle indexes podcasts alongside web stories so readers can play short audio clips linked to related news items.
- The app uses embedding models to match podcast segments to news stories and AI to decide precise clip start and end points.
Podcasts Are Becoming Primary News Channels
- Podcasts are becoming a primary venue for breaking news and public figures' announcements, increasing their news value.
- Tech CEOs now favor friendly podcast hosts to push talking points, making podcast monitoring essential for up-to-date reporting.
Embedding Models Power Semantic Matching Not Generation
- Particle uses vector embeddings (from companies that also provide LLMs) not to generate text but to semantically link podcast segments to stories.
- Transcription comes from 11 Labs while Particle's clipping logic remains proprietary.