Introducing real-time video tagging and speech-to-text


We've been working with the world's top media companies to automatically tag their video assets for discoverability and higher engagement. They rely on us to extract not only what appears or is mentioned, but most importantly what is relevant about their content in order to really understand it. This level of understanding is how we outperform even the tech giants when it comes to content metadata. When building a system for real-time scenarios such as live streaming, our partners were very clear that it wasn't just about speed but also quality; real-time understanding, not tagging. So over the past few months, we've been redesigning our solution with this goal: reducing as much latency as possible while keeping our core focus on relevancy.