Hey! We're not using text-to-speech, all of the narrations we have so far (around 4000 audio articles) are recorded by people. We're currently doing a mix of both news and content that's more on the evergreen side (science, books, startups, self-dev). The aim here is to basically create a huge library of audio content on a wide range of topics and be able to make recommendations to users, based on what they're usually interested in. Thanks a lot for the feedback, it's very highly appreciated!