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
Journal ArticleDOI
Discover, Explanation, Improvement: Automatic Slice Detection Framework for Natural Language Processing
TL;DR: In this article , the authors propose a discover, explanation, improvement (DEI) framework that discovers coherent and underperforming groups of data points and unifies data points of each slice under human-understandable concepts.
Proceedings ArticleDOI
TriNet: stabilizing self-supervised learning from complete or slow collapse
TL;DR: TriNet as discussed by the authors introduces a triple-branch architecture for preventing collapse and stabilizing the pre-training, which achieves a relative word error rate reduction (WERR) of 6.06% compared to the state-of-the-art Data2vec.
Journal ArticleDOI
Deep Learning for Joint Acoustic Echo and Acoustic Howling Suppression in Hybrid Meetings
Hao Zhang,Meng Yu,Dong Yu +2 more
TL;DR: In this paper , a self-attentive recurrent neural network is utilized to extract the target speech from microphone recordings with accessible and learned reference signals, thus suppressing acoustic echo and acoustic howling simultaneously.
Posted Content
Prediction-Adaptation-Correction Recurrent Neural Networks for Low-Resource Language Speech Recognition
TL;DR: In this paper, a prediction-adaptation-correction recurrent neural networks (PAC-RNNs) is proposed for low-resource speech recognition, which is comprised of a pair of neural networks in which a correction network uses auxiliary information given by a prediction network to help estimate the state probability.
Patent
Reconnaissance de parole mixte
TL;DR: In this article, a system for the reconnaissance de parole mixte provenant d'une source is presented, which concerne un systeme and un procede for the reconnaitre le signal de parole parle par l'orateur avec un niveau plus faible de la caracteristique de parle a partir de l'echantillon de Parole mixte.