专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US07502736B2 Voice registration method and system, and voice recognition method and system based on voice registration method and system 有权
标题翻译：语音注册方法和系统，语音识别方法和系统基于语音注册的方法和系统
公开(公告)号：US07502736B2
公开(公告)日：2009-03-10
申请号：US10486258
申请日：2001-12-06
申请人： Sang-jin Hong , Sung-zoo Lee , Tae-soo Kim , Tae-sung Lee , Ho-jin Choi , Byoung-won Hwang
发明人： Sang-jin Hong , Sung-zoo Lee , Tae-soo Kim , Tae-sung Lee , Ho-jin Choi , Byoung-won Hwang
IPC分类号： G10L15/06 , G10L15/16 , G10L17/00
CPC分类号： G10L15/063
摘要： Disclosed is a voice registration method for voice recognition, comprising the steps of analyzing a spectrum of a sound signal inputted from the outside; extracting predetermined language units for a speaker recognition from a voice signal in the sound signal; measuring the loudness of each language unit; collecting voice data on registered (background) speakers including loudness data of the plurality of background speakers as a reference onto voice database; determining whether the loudness of each language unit is within a predetermined loudness range based on the voice data base; learning each language unit by using a multi-layer perceptron in the case that at least a predetermined number of language units are within the predetermined loudness range; and storing data on the learned language unit as data for recognizing the speaker. With this configuration, loudness of a speaker is considered at learning for registering his/her voice and at verifying a speaker.
摘要翻译：公开了一种用于语音识别的语音注册方法，包括以下步骤：分析从外部输入的声音信号的频谱; 从声音信号中的语音信号中提取用于说话者识别的预定语言单元; 测量每个语言单位的响度; 收集多个背景扬声器的响度数据的登记（背景）扬声器上的语音数据作为参考语音数据库; 基于所述语音数据库确定每个语言单元的响度是否在预定响度范围内; 在至少预定数量的语言单位在预定响度范围内的情况下，通过使用多层感知器来学习每个语言单元; 并且将所学习的语言单元上的数据存储为用于识别说话者的数据。利用这种配置，在学习用于注册他/她的声音和在验证扬声器时考虑扬声器的响度。

2. 发明申请

US20050033573A1 Voice registration method and system, and voice recognition method and system based on voice registration method and system 有权
标题翻译：语音注册方法和系统，语音识别方法和系统基于语音注册的方法和系统
公开(公告)号：US20050033573A1
公开(公告)日：2005-02-10
申请号：US10486258
申请日：2001-12-06
申请人： Sang-jin Hong , Sung-zoo Lee , Tae-soo Kim , Tae-sung Lee , Byoung-won Hwang
发明人： Sang-jin Hong , Sung-zoo Lee , Tae-soo Kim , Tae-sung Lee , Byoung-won Hwang
IPC分类号： G06N3/00 , G10L15/00 , G10L15/06 , G10L15/16 , G10L17/00
CPC分类号： G10L15/063
摘要： Disclosed is a voice registration method for voice recognition, comprising the steps of analyzing a spectrum of a sound signal inputted from the outside; extracting predetermined language units for a speaker recognition from a voice signal in the sound signal; measuring the loudness of each language unit; collecting voice data on registered(background) speakers including loudness data of the plurality of background speakers as a reference onto voice database; determining whether the loudness of each language unit is within a predetermined loudness range based on the voice data base; learning each language unit by using a multi-layer perceptron in the case that at least a predetermined number of language units arc within the predetermined loudness range; and storing data on the learned language unit as data for recognizing the speaker. With this configuration, loudness of a speaker is considered at learning for registering his/her voice and at verifying a speaker, so that it is possible to more correctly verify the speaker.
摘要翻译：公开了一种用于语音识别的语音注册方法，包括以下步骤：分析从外部输入的声音信号的频谱; 从声音信号中的语音信号中提取用于说话者识别的预定语言单元; 测量每个语言单位的响度; 收集多个背景扬声器的响度数据的登记（背景）扬声器上的语音数据作为参考语音数据库; 基于所述语音数据库确定每个语言单元的响度是否在预定响度范围内; 在至少预定数量的语言单位处于预定响度范围内的情况下，通过使用多层感知器来学习每个语言单元; 并且将所学习的语言单元上的数据存储为用于识别说话者的数据。利用这种配置，在学习用于注册他/她的声音和在验证扬声器时考虑扬声器的响度，使得可以更正确地验证扬声器。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式