Deep Learning and its Application on Speech Processing

Abstract

By establishing new state-of-the-art performance in speech recognition, image recognition, and some natural language processing tasks, deep learning techniques have achieved tremendous successes in recent years. In this talk, I will give a brief introduction on deep learning technology, and give three examples for its application on speech processing: (1) Machine listens to and understands audio content, and answers the related questions. (2) Agent interacts with users by deep reinforcement learning (which is the technique also used in Alpha Go). (3) Machine listens to hours of audio books and tries to learn human languages without any supervision.