Bahareh Tolooshams

I received my B.A.Sc. degree (Hons.) in electrical engineering from the University of Waterloo in 2017.  I am currently pursuing my Ph.D. in electrical engineering with the CRISP Group at Harvard University.  My research interests lie at the intersection of statistical learning theory, signal processing, and computational neuroscience.  Specifically, I study behavior and learning dynamics of model-based deep learning and its application to inverse problems.

I spent the summer of 2019 at AWS, Amazon AI, as an Applied Scientist Intern working on neural architectures for multichannel speech enhancement, and I am part of the Applied Sciences Group (ASG) at Microsoft as a research intern for summer, 2021, working on stereo speech enhancement.  To know more about my research, see https://btolooshams.github.io.

I enjoy outdoor activities such as biking and reading books.

• A Training Framework for Stereo-Aware Speech Enhancement Using Deep Neural Networks

  Proceedings, ICASSP 2022 May, 2022 Pages 6962-6966

  By Bahareh Tolooshams, Kazuhito Koishida