Xiaohui Zhang

Xiaohui is a Research Scientist and one of the founding members of the Video Speech Recognition team at Facebook AI Applied Research. Xiaohui received an M.S.E. in Applied Mathematics and Statistics and a Ph.D. from the Center for Language and Speech Processing at the Johns Hopkins University. Xiaohui's areas of interest include acoustic modeling, pronunciation learning and OOV recovery for ASR, 2nd order optimization methods for deep learning, and actively contributed to the open-source ASR toolkit Kaldi.

Xiaohui's Publications

RESEARCH

NLP

From Senones to Chenones: Tied Context-Dependent Graphemes for Hybrid Speech Recognition

There is an implicit assumption that traditional hybrid approaches for automatic speech recognition (ASR) cannot directly model graphemes and need to rely on phonetic lexicons to get competitive performance, especially on English which has poor…

Duc Le, Xiaohui Zhang, Weiyi Zhang, Christian Fuegen, Geoffrey Zweig, Michael L. Seltzer,