CASIA OpenIR  > 模式识别国家重点实验室  > 语音交互
Adversarial Multilingual Training for Low-resource Speech Recognition
Yi JY(易江燕)1,2; Tao Jianhua1,2,3; Wen Zhengqi1; Bai Ye1,2
2018-04-15
会议名称2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2018)
页码4899-4903
会议日期15–20 April, 2018
会议地点Calgary, Alberta, Canada
摘要
This paper proposes an adversarial multilingual training to train bottleneck (BN) networks for the target language. A parallel shared-exclusive model is also proposed to train the BN network. Adversarial training is used to ensure that the shared layers can learn language-invariant features. Experiments are conducted on IARPA Babel datasets. The results show that the proposed adversarial multilingual BN model outperforms the baseline BN model by up to 8.9% relative word error rate (WER) reduction. The results also show that the proposed parallel shared-exclusive model achieves up to 1.7% relative WER reduction when compared
with the stacked share-exclusive model.
关键词Speech Recognition Low-resource Deep Neural Networks Bottleneck Features Adversarial Multilingual Training
收录类别EI
语种英语
文献类型会议论文
条目标识符http://ir.ia.ac.cn/handle/173211/20908
专题模式识别国家重点实验室_语音交互
作者单位1.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Science
2.School of Artificial Intelligence, University of Chinese Academy of Sciences
3.CAS Center for Excellence in Brain Science and Intelligence Technology, Institute of Automation, Chinese Academy of Sciences
推荐引用方式
GB/T 7714
Yi JY,Tao Jianhua,Wen Zhengqi,et al. Adversarial Multilingual Training for Low-resource Speech Recognition[C],2018:4899-4903.
条目包含的文件
条目无相关文件。
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[Yi JY(易江燕)]的文章
[Tao Jianhua]的文章
[Wen Zhengqi]的文章
百度学术
百度学术中相似的文章
[Yi JY(易江燕)]的文章
[Tao Jianhua]的文章
[Wen Zhengqi]的文章
必应学术
必应学术中相似的文章
[Yi JY(易江燕)]的文章
[Tao Jianhua]的文章
[Wen Zhengqi]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。