D
Dong Yu
Researcher at Tencent
Publications - 389
Citations - 45733
Dong Yu is an academic researcher from Tencent. The author has contributed to research in topics: Artificial neural network & Word error rate. The author has an hindex of 72, co-authored 339 publications receiving 39098 citations. Previous affiliations of Dong Yu include Peking University & Microsoft.
Papers
More filters
Book ChapterDOI
Training and Decoding Speedup
TL;DR: This chapter describes the parallel training algorithms such as pipelined backpropagation algorithm, asynchronous stochastic gradient descend algorithm, and augmented Lagrange multiplier algorithm and introduces model size reduction techniques based on low-rank approximation which can speedup both training and decoding.
NeuralKalman: A Learnable Kalman Filter for Acoustic Echo Cancellation
TL;DR: In this article , the authors integrate the frequency domain Kalman filter (FDKF) and deep neural networks (DNNs) into a hybrid method, called NeuralKalman, to leverage the advantages of deep learning and adaptive filtering algorithms.
Journal ArticleDOI
Cooperative Fault-Tolerant Control for a Class of Nonlinear MASs by Resilient Learning Approach.
TL;DR: In this paper , a learning-based resilient fault-tolerant control method is proposed for a class of uncertain nonlinear multiagent systems (MASs) to enhance the security and reliability against denial-of-service (DoS) attacks and actuator faults.
Patent
Configurable grammar templates
TL;DR: In this article, grammar extensions are provided that allow application developers to selectively include portions of grammar templates and to easily combine grammar elements to form new grammar structures to provide application developers with the ability to easily form customized grammars.
Posted Content
A Comparison of Lattice-free Discriminative Training Criteria for Purely Sequence-Trained Neural Network Acoustic Models
Chao Weng,Dong Yu +1 more
TL;DR: It is demonstrated that, analogous to LF-MMI, a neural network acoustic model can also be trained from scratch using LF-bMMI or LF-sMBR criteria respectively without the need of cross-entropy pre-training.