Dr. Zhong-Qiu Wang
Postdoctoral Research Associate, Carnegie Mellon University
Joint ECE/CS Faculty Candidate Seminar
Wednesday, April 12, 2023, 9:00-10:00am
B02 CSL and Online via Zoom
Title: Deep learning based speech separation
Abstract: Voice-controlled devices such as Amazon Echo and Google Home have become very popular in recent years. In realistic conditions, microphones usually capture a mixture of target speech and interference signals, which can be a combination of environmental noises, room reverberation, and concurrent speech by other speakers. The interferences are very detrimental to speech processing and human hearing. In this talk, I will introduce my work on deep learning based supervised and unsupervised speech separation, where deep neural networks are designed to enhance target speech, reduce noises and reverberation, and separate concurrent speech by multiple speakers into individual speaker signals, based on a single microphone or an array of microphones.
Zhong-Qiu Wang is currently a postdoctoral research associate in the Language Technologies Institute at Carnegie Mellon University (CMU). He obtained his Ph.D. degree in computer science from The Ohio State University in 2020. He was a visiting research scientist at Mitsubishi Electric Research Laboratories from 2020 to 2021. His research interests include computer audition, speech separation, microphone array processing, robust automatic speech recognition, and deep learning. He won a Best Student Paper Award at ICASSP 2018. He is a member of the IEEE Audio and Acoustic Signal Processing Technical Committee (AASP-TC). You can learn more about his research at http://zqwang7.github.io/.