CASIA OpenIR  > 数字内容技术与服务研究中心  > 听觉模型与认知计算
Multidimensional Residual Learning Based on Recurrent Neural Networks for Acoustic Modeling
Zhao, Yuanyuan; Xu, Shuang; Xu, Bo; Yuanyuan Zhao
2016-09
Conference NameInterspeech2016
Pages3419-3423
Conference DateSeptember 8-12
Conference PlaceSan Francisco, USA
AbstractTheoretical and empirical evidences indicate that the depth of neural networks is crucial to acoustic modeling in speech recognition tasks. Unfortunately, the situation in practice always is that with the depth increasing, the accuracy gets saturated and then degrades rapidly.
    In this paper, a novel multidimensional residual learning architecture is proposed to address this degradation of deep recurrent neural networks (RNNs) on acoustic modeling by further exploring the spatial and temporal dimensions. In the spatial dimension, shortcut connections are introduced to RNNs, along which the information can flow across several layers without attenuation. In the temporal dimension, we cope with the degradation problem by regulating temporal granularity, namely, splitting the input sequence into several parallel sub-sequences, which can ensure information flowing across the time axis unimpededly. Finally, we place a row convolution layer on the top of all recurrent layers to comprehend appropriate information from several parallel sub-sequences to feed to the classifier. Experiments are illustrated on two quite different speech recognition tasks and 10% relative performance improvements are observed.
KeywordAcoustic Modeling Multidimensional Residual Learning Long Short-term Memory Block Row Convolution Layer
Indexed ByEI
Language英语
Document Type会议论文
Identifierhttp://ir.ia.ac.cn/handle/173211/19648
Collection数字内容技术与服务研究中心_听觉模型与认知计算
Corresponding AuthorYuanyuan Zhao
AffiliationInstitute of Automation, Chinese Academy of Sciences
Recommended Citation
GB/T 7714
Zhao, Yuanyuan,Xu, Shuang,Xu, Bo,et al. Multidimensional Residual Learning Based on Recurrent Neural Networks for Acoustic Modeling[C],2016:3419-3423.
Files in This Item: Download All
File Name/Size DocType Version Access License
Multidimensional Res(298KB)会议论文 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zhao, Yuanyuan]'s Articles
[Xu, Shuang]'s Articles
[Xu, Bo]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zhao, Yuanyuan]'s Articles
[Xu, Shuang]'s Articles
[Xu, Bo]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zhao, Yuanyuan]'s Articles
[Xu, Shuang]'s Articles
[Xu, Bo]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Multidimensional Residual Learning Based on Recurrent Neural Networks for Acoustic Modeling.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.