Emformer: Efficient Memory Transformer Based Acoustic Model for Low Latency Streaming Speech Recognition
Yangyang Shi, Yongqiang Wang, Chunyang Wu, Ching-Feng Yeh, Julian Chan, Frank Zhang, Duc Le, Mike Seltzer
People
Chunyang Wu
Research Scientist
I joined Meta in September 2019 as a Research Scientist. I am currently conducting the research and development of spoken language technology for Meta community. Prior to joining Meta, I did my PhD at the University of Cambridge. I am interested in machine learning in general.
Interests
Machine learning, natural language processing & speech
Latest Publications
ICASSP - 2021
Dynamic Encoder Transducer: A Flexible Solution For Trading Off Accuracy For Latency
Yangyang Shi, Varun Nagaraja, Chunyang Wu, Jay Mahadeokar, Duc Le, Rohit Prabhavalkar, Alex Xiao, Ching-Feng Yeh, Julian Chan, Christian Fuegen, Ozlem Kalinli, Michael L. Seltzer
Transformer-based Acoustic Modeling for Streaming Speech Synthesis
Chunyang Wu, Zhiping Xiu, Yangyang Shi, Ozlem Kalinli, Christian Fuegen, Thilo Koehler, Qing He
Weak-Attention Suppression For Transformer Based Speech Recognition
Yangyang Shi, Yongqiang Wang, Chunyang Wu, Christian Fuegen, Frank Zhang, Duc Le, Ching-Feng Yeh, Michael L. Seltzer