CASIA OpenIR

Browse/Search Results:  1-10 of 41 Help

Selected(0)Clear Items/Page:    Sort:
Speech emotion recognition from variable-length inputs with triplet loss function 会议论文
Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, India, 2018
Authors:  Huang, Jian;  Li, Ya;  Tao, Jianhua;  Lian, Zheng
Favorite  |  View/Download:22/0  |  Submit date:2018/11/26
Syllable-Based Sequence-to-Sequence Speech Recognition with the Transformer in Mandarin Chinese 会议论文
Interspeech, 印度的海德拉巴, 2018
Authors:  Shiyu Zhou;  Linhao Dong;  Shuang Xu;  Bo Xu
View  |  Adobe PDF(416Kb)  |  Favorite  |  View/Download:43/9  |  Submit date:2018/12/20
Asr  Multi-head Attention  Syllable Based Acoustic Modeling  Sequence-to-sequence  
Multilingual Recurrent Neural Networks with Residual Learning for Low-Resource Speech Recognition 会议论文
Interspeech, Stockholm, 2017
Authors:  Shiyu Zhou;  Yuanyuan Zhao;  Shuang Xu;  Bo Xu
View  |  Adobe PDF(269Kb)  |  Favorite  |  View/Download:102/16  |  Submit date:2017/09/20
Lstm  Multilingual Speech Recognition  Low-resource  Residual Learning  Shared-hidden-layer  
End-to-end Language Identification using Attention-based Recurrent Neural Networks 会议论文
InterSpeech2016, San Francisco, USA, 2016.9.8-2016.9.12
Authors:  Wang Geng;  Wenfu Wang;  Yuanyuan Zhao;  Xinyuan Cai;  Bo Xu;  Cai Xinyuan
View  |  Adobe PDF(610Kb)  |  Favorite  |  View/Download:122/30  |  Submit date:2016/10/28
Language Identification  End-to-end Training  Attention  
The Parameterized Phoneme Identity Feature as a Continuous Real-Valued Vector for Neural Network based Speech Synthesis 会议论文
INTERSPEECH, San Francisco,USA, Sep 8-12, 2016
Authors:  Wen ZQ(温正棋);  Li Y(李雅);  Tao JH(陶建华);  Wen, Zhengqi
View  |  Adobe PDF(541Kb)  |  Favorite  |  View/Download:69/11  |  Submit date:2016/10/28
Phoneme Embedded Vector  Word Embedding  Speech Synthesis  Blstm-rnn  
Gating Recurrent Enhanced Memory Neural Networks on Language Identification 会议论文
InterSpeech 2016, San Francisco, USA, 2016.9.8-2016.9.12
Authors:  Wang Geng;  Yuanyan Zhao;  Wenfu Wang;  Cai Xinyuan;  Bo Xu;  Xinyuan Cai
View  |  Adobe PDF(529Kb)  |  Favorite  |  View/Download:83/23  |  Submit date:2016/10/28
Language Identification  Gating Recurrent Neural Networks  Learnable Enhanced Memory Block  Sortagrad-like Training Approach  Hard Sample Score Acquisition  
Multilingual Tandem Bottleneck Feature For Language Identification 会议论文
Interspeech2015, Dresden,German, 2016.9.6-2016.9.10
Authors:  Wang Geng;  Jie Li,;  Shanshan Zhang;  Xinyuan Cai;  Bo Xu;  Xinyuan.Cai
View  |  Adobe PDF(529Kb)  |  Favorite  |  View/Download:80/14  |  Submit date:2016/10/28
Language Identification  Deep Bottleneck Feature  Tandem Feature  Multi-deep Feature  Multi-training Procedure.  
Towards End-to-End Speech Recognition for Chinese Mandarin using Long Short-Term Memory Recurrent Neural Networks 会议论文
Interspeech 2015, Dersen,German, 2016.9.6-2016.9.10
Authors:  Jie Li;  Heng Zhang;  Xinyuan Cai;  Bo Xu
View  |  Adobe PDF(531Kb)  |  Favorite  |  View/Download:107/29  |  Submit date:2016/10/28
Long Short-term Memory  End-to-end  Connectionist Temporal Classification  Speech Recognition  
Improving Large Vocabulary Accented Mandarin Speech Recognition with Attribute-based I-vectors 会议论文
INTERSPEECH, San Francisco, America, 2016
Authors:  Hao Zheng;  Shanshan Zhang;  Liwei Qiao;  Jianping Li;  Wenju Liu
View  |  Adobe PDF(206Kb)  |  Favorite  |  View/Download:133/14  |  Submit date:2016/06/28
Accented Speech Recognition  Large Vocabulary Continuous Speech Recognition  Attribute  
Towards end-to-end speech recognition for Chinese Mandarin using long short-term memory recurrent neural networks 会议论文
INTERSPEECH 2015 emphasizes an interdisciplinary approach covering all aspects of speech science and technology spanning basic theories to applications. In addition to regular oral and poster sessions, the conference will also feature plenary talks by internationally renowned experts, tutorials, special sessions, show & tell sessions, and exhibits. A number of satellite events will take place immediately before and after the conference., 德国, 2015年
Authors:  Li J(李杰);  Zhang H(张恒);  Cai XY(蔡新元);  Xu B(徐波)
View  |  Adobe PDF(504Kb)  |  Favorite  |  View/Download:181/42  |  Submit date:2016/04/11