Getting Your TV to Understand You

Its  the ultimate challenge, understanding complex context.  Good direction.

Understanding in Context   University of Waterloo News

Researchers at the University of Waterloo in Canada, the University of Maryland, and the Comcast Applied AI Research Laboratory have developed a method to improve the voice query understanding capabilities of home entertainment platforms. The method relies on artificial intelligence (AI) technology to achieve the most natural speech-based interaction with TVs to date. The researchers utilized a technique known as hierarchical recurrent neural networks to better model context and improve the system's accuracy. In January, the researchers' neural network model was deployed in production to answer queries from live users, and they found that it handles the majority of highly complicated queries appropriately, greatly enhancing user experience. Waterloo's Jimmy Lin said, "What's special about this approach is that we take advantage of context—such as previously watched shows and favorite channels—to personalize results, thereby increasing accuracy."and is to shoppers in a broader category. ... "