摘要详情
  1. 活动首页
  2. 摘要清单
  3. 摘要详情
37 / 2021-06-22 21:39:39
Pathological Voice Detection Using Transfer Learning Methods
Biomedical Measurements
终稿
Yihua Zhang / Soochow University
Xincheng Zhu / Soochow University
Yuanbo Wu / Soochow University
xiaojun Zhang / Soochow University
Yishen Xu / Soochow University
Zhi Tao / Soochow University
 



Abstract—Pathological voice detection has obtained great progress on recognition rate. However, these results are all achieved in single corpus. In cross-corpus detection, the accuracy of detection tends to drop. Transfer learning, which has been proven to be able to address cross-domain recognition, is applied to pathological voice detection. The aim of transfer learning is to find a mapping matrix, which maps the source and target data into a common feature subspace to reduce the difference of the feature distribution between different corpora. Then a classifier trained on source domain will be applied to the target domain without labels. Meanwhile inner-class and inter-class distance are applied to transfer learning to keep the features of the original space as much as possible after dimensionality reduction. Three-dimensional scatter plots of top 3 features for all samples in source and target corpora are presented to demonstrate the effectiveness of the algorithm. Finally, after transfer learning methods, the accuracy is increased by up to 9.14% . Experimental results demonstrate that the accuracy of cross-corpus pathological voice detection can signification improve by transfer learning methods.

 
重要日期

  • 会议日期

    10月21日

    2021

    至

    10月23日

    2021

  • 10月26日 2021

    注册截止日期

主办单位
Southeast University, China
联系方式
登录查看完整联系方式
历届会议
移动端
在手机上打开
小程序
打开微信小程序
客服
扫码或点此咨询