An AI news app called Particle, developed by a former Twitter engineer, can now keep track of not only news published on the web, but also breaking news in podcasts.
Ahead of its recent Android release, Particle introduced a feature called Podcast Clips. It finds the most interesting and relevant moments from different types of podcasts and includes those clips in your feed along with related news articles.
So instead of listening to a long podcast just to hear 45 seconds of interesting commentary, you can play clips while reading the news on Particle. You also have the option to read a transcript of the clip instead, as the words being spoken are highlighted.
“We’ve been doing that for basically every news story. If there’s a podcast talking about it or if it’s relevant, we have all those clips,” Particle CEO Sara Beykpour, who previously served as Twitter’s senior director of product management, told TechCrunch. “It’s such a great way to be able to see what people are saying about this and what their commentary is when you read a story or learn about a story.”
This addition acknowledges the changes in the news ecosystem that have been underway for years. Not only are more people getting their news from podcasts and trusting them as a reliable source of information, but podcasts are also becoming a destination for breaking news and important announcements from celebrities.
As Bloomberg reported in 2024, tech company CEOs in particular are now looking for friendly podcast hosts to air their talking points, rather than trying to work with traditional media.
That makes it even more important to pay attention to podcasts to keep up with the news.
Beykpour said Particle uses an embedding model to understand when a podcast is relevant to a particular news article. Although these models are provided by the same companies that offer LLM models, they are not generative AI technologies, she explains.
“We use vector embeddings to understand how different parts of the podcast relate to different stories,” Beikpour said. “One podcast might cover 10 or 20 stories, so we use AI to figure that out. We also use AI to do some of the logic around clipping and understand when clips start and end.”
The company uses Eleven Lab technology for transcription. But the piece of technology that determines exactly where to clip audio is part of Particle’s secret sauce.
The idea of using podcasts to provide a deeper understanding of news commentary has also been explored in newsrooms recently. As Nieman Institute reported this month, the New York Times is using a custom AI tool powered by LLM to transcribe and summarize new episodes of dozens of right-wing and more conservative podcasts to better understand what influencers on that side are saying about the news.
Particle’s Podcast Clips feature isn’t just tied to news articles. Since the app already understands different entities such as people, places, and things, you can go to a prominent person’s page like OpenAI CEO Sam Altman and see all of his appearances on podcasts placed as a feed.
Particle is also hard at work building other features. The company is making its first attempt at monetization with Particle+, an optional subscription for $2.99 per month (or $29.99 per year) that gives you access to premium features. These include the ability to summarize the news in your preferred style using natural language. Choose from a variety of voices when using your personalized audio feed. “Listen to the News”; unlimited crossword puzzles. AI chatbot support for private questions. And so on.
There are several other notable changes in the Android release. In addition to the typical sections like Politics, Technology, and Entertainment, the Browse tab now includes timely articles like the 2026 Winter Olympics. Additionally, tapping on an entity will bring up a new page with definitions, stories, articles, related entities, and related topics.
Particle doesn’t share data on user activity or conversion rates, but Beykpour pointed to the app’s global audience before Android. On a weekly basis, 55% of Particle users are outside the US, with India (15%) being the second largest market after the US.
Source link
