Even for humans, finding answers to questions that involve high-level concepts is sometimes difficult. So how how can we use AI to search 8,000 hours of video content and return meaningful answers? In this video, Matt Coatney explains how cutting-edge technologies like video processing, voice to text, and concept extraction combine to make video content more searchable. Learn how you can open up non-text content to search with deep learning.

Continue exploring deep learning on Safari with the full Business of Deep Learning course.


Matt Coatney is a technology executive, business advisor, entrepreneur, author, and speaker. His focus is on bringing advanced artificial intelligence and analytic technologies to market. He has co-founded three companies, advised several others, and contributed to the early success of two different tech startups. Matt has also launched data analytics products designed for the fields of life sciences, healthcare, government, finance, and law. Currently Matt is the VP of Services for Exaptive; he previously was the IT strategy lead for global law firm WilmerHale, and was in charge of technology and operations for a legal search product at LexisNexis. Matt writes and speaks frequently on business and advanced technology topics, and his work has appeared at TEDx and other international conferences, as well as in webinars, books and technical journals.

Article image: Screen from "How does IBM Watson search TED Talks?" (source: O'Reilly).