首页> 外国专利> DEEPLEARING METHOD FOR VOICE RECOGNITION MODEL AND VOICE RECOGNITION DEVICE BASED ON ARTIFICAL NEURAL NETWORK

DEEPLEARING METHOD FOR VOICE RECOGNITION MODEL AND VOICE RECOGNITION DEVICE BASED ON ARTIFICAL NEURAL NETWORK

机译:基于人工神经网络的语音识别模型和语音识别装置的深度学习方法

摘要

A method for training a speech recognition model based on an artificial neural network is disclosed. A method for training a speech recognition model based on an artificial neural network according to an embodiment of the present invention includes target data representing a user's speech feature and non-target data representing non-features representing noise other than the user's speech (non-target data). -target data) as input and output values at random to learn the user's voice, and is a voice recognition model that recognizes the user's voice in a situation containing noise. The method and voice recognition device for learning an artificial neural network-based voice recognition model of the present invention include an artificial intelligence module, a robot, an augmented reality (AR) device, a virtual reality (VR) device, and 5G It may be associated with a device related to a service, and the like.
机译:公开了一种基于人工神经网络训练语音识别模型的方法。 根据本发明实施例的用于训练基于人工神经网络的语音识别模型的方法包括表示用户语音特征和表示表示除用户语音之外的噪声的非特征的非目标数据(非目标 数据)。 -target数据)作为随机的输入和输出值来学习用户的语音,并且是一种语音识别模型,可以在包含噪声的情况下识别用户的语音。 用于学习本发明的人工神经网络的语音识别模型的方法和语音识别装置包括人工智能模块,机器人,增强现实(AR)设备,虚拟现实(VR)设备和5G 与与服务相关的设备等相关联。

著录项

  • 公开/公告号KR102321798B1

    专利类型

  • 公开/公告日2021-11-05

    原文格式PDF

  • 申请/专利权人 엘지전자 주식회사;

    申请/专利号KR20190099958

  • 发明设计人 김다미;

    申请日2019-08-15

  • 分类号G10L15/06;H04L5;H04W72/04;

  • 国家 KR

  • 入库时间 2022-08-24 22:27:57

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