Sheng Li

About

Sheng Li is an Assistant Professor at the Institute of Science Tokyo, Japan. His research areas include multilingual automatic speech recognition (ASR), speech separation techniques, and dynamic data augmentation, as reflected in his recent articles. He has contributed to advancements in language capabilities for ASR models, as well as exploring low-resource settings and indigenous languages such as Tibetan. His work also encompasses challenges in speech recognition, including disordered speech and adversarial attacks on dialogue systems.