Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Xiaohui Zhang

Fast, Simpler and More Accurate Hybrid ASR Systems Using Wordpieces

May 19, 2020
Frank Zhang, Yongqiang Wang, Xiaohui Zhang, Chunxi Liu, Yatharth Saraf, Geoffrey Zweig

* submitted to interspeech 2020 

  Access Paper or Ask Questions

Deja-vu: Double Feature Presentation in Deep Transformer Networks

Oct 23, 2019
Andros Tjandra, Chunxi Liu, Frank Zhang, Xiaohui Zhang, Yongqiang Wang, Gabriel Synnaeve, Satoshi Nakamura, Geoffrey Zweig

  Access Paper or Ask Questions

Transformer-based Acoustic Modeling for Hybrid Speech Recognition

Oct 22, 2019
Yongqiang Wang, Abdelrahman Mohamed, Duc Le, Chunxi Liu, Alex Xiao, Jay Mahadeokar, Hongzhao Huang, Andros Tjandra, Xiaohui Zhang, Frank Zhang, Christian Fuegen, Geoffrey Zweig, Michael L. Seltzer

  Access Paper or Ask Questions

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

Oct 11, 2019
Duc Le, Xiaohui Zhang, Weiyi Zheng, Christian Fügen, Geoffrey Zweig, Michael L. Seltzer

* To appear at ASRU 2019 

  Access Paper or Ask Questions

Multilingual ASR with Massive Data Augmentation

Sep 14, 2019
Chunxi Liu, Qiaochu Zhang, Xiaohui Zhang, Kritika Singh, Yatharth Saraf, Geoffrey Zweig

  Access Paper or Ask Questions

Automated vehicle's behavior decision making using deep reinforcement learning and high-fidelity simulation environment

Apr 17, 2018
Yingjun Ye, Xiaohui Zhang, Jian Sun

* 22 pages, 13 figures, CICTP2018 

  Access Paper or Ask Questions

Acoustic data-driven lexicon learning based on a greedy pronunciation selection framework

Jun 12, 2017
Xiaohui Zhang, Vimal Manohar, Daniel Povey, Sanjeev Khudanpur

  Access Paper or Ask Questions

Parallel training of DNNs with Natural Gradient and Parameter Averaging

Jun 22, 2015
Daniel Povey, Xiaohui Zhang, Sanjeev Khudanpur

* Accepted as workshop contribution to ICLR 2015. 12 pages plus 16 pages of appendices, International Conference on Learning Representations (ICLR): Workshop track, 2015. [2 sets of minor fixes post-publication.] 

  Access Paper or Ask Questions