Finding a new podcast you actually like is easier said than done. It’s a task that requires time, effort, and a good chunk of your attention. Spotify wants to make this tedious process easier and is betting that podcast discovery technology developed by a startup will help its users find and get hooked on new shows.
The company announced on Thursday that it had acquired the podcast discovery platform Podz, a startup that developed a so-called “audio newsfeed” that features 60-second clips from different podcasts. Here’s where the technology comes in: Instead of simply reproducing the clips chosen by podcast creators, Podz uses a machine learning model to select clips of shows’ best moments. This allows users to preview a variety of podcasts in minutes rather than listening to entire 30- or 60-minute shows.
Spotify said that combining Podz with its 2.6 million podcast repertoire, learnings from its previous work in music discovery, and current investments in podcast recommendation will allow it to take podcast discovery to the next level. Altogether, the company maintains that it will be easier for users to find shows they like and for podcast creators to build a fan base.
“Spotify has had machine learning experts focused on improving audio discovery for almost a decade, but there is more work to be done,” the company said in a news announcement. “We believe that Podz’ technology will complement and accelerate Spotify’s focused efforts to drive discovery, deliver listeners the right content at the right time, and accelerate growth of the category worldwide.”
According to TechCrunch, Podz’s machine learning model was trained on more than 100,000 hours of audio and considered input from journalists and audio editors. Back in February, the outlet reported that Podz had raised $2.5 million in pre-seed funding. Besides receiving investments from venture capital firms like M13 and Canaan Partners, it also had backers like Katie Couric and Paris Hilton. (For reference, both Couric and Hilton have podcasts).