首页> 外国专利> Adversarial teacher-student learning for unsupervised domain adaptation

Adversarial teacher-student learning for unsupervised domain adaptation

机译:对抗性师生的无监督领域适应

摘要

Methods, systems, and computer programs are presented for training, with adversarial constraints, a student model for speech recognition based on a teacher model. One method includes operations for training a teacher model based on teacher speech data, initializing a student model with parameters obtained from the teacher model, and training the student model with adversarial teacher-student learning based on the teacher speech data and student speech data. Training the student model with adversarial teacher-student learning further includes minimizing a teacher-student loss that measures a divergence of outputs between the teacher model and the student model; minimizing a classifier condition loss with respect to parameters of a condition classifier; and maximizing the classifier condition loss with respect to parameters of a feature extractor. The classifier condition loss measures errors caused by acoustic condition classification. Further, speech is recognized with the trained student model.
机译:提出了用于在对抗性约束下训练基于教师模型的语音识别学生模型的方法,系统和计算机程序。一种方法包括以下操作:基于教师语音数据训练教师模型;使用从教师模型获得的参数初始化学生模型;以及基于教师语音数据和学生语音数据通过对抗性师生学习训练学生模型。用对抗性师生学习训练学生模型还包括最小化师生损失,该损失可衡量教师模型和学生模型之间的输出差异;关于条件分类器的参数,最小化分类器条件损失;以及相对于特征提取器的参数最大化分类器条件损失。分类器条件损失测量由声学条件分类引起的误差。此外,通过训练有素的学生模型可以识别语音。

著录项

  • 公开/公告号US10643602B2

    专利类型

  • 公开/公告日2020-05-05

    原文格式PDF

  • 申请/专利权人 MICROSOFT TECHNOLOGY LICENSING LLC;

    申请/专利号US201815923795

  • 发明设计人 JINYU LI;ZHONG MENG;YIFAN GONG;

    申请日2018-03-16

  • 分类号G10L15/06;G06N20;G10L15/16;G10L15/02;

  • 国家 US

  • 入库时间 2022-08-21 11:26:35

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号