-
公开(公告)号：US20140114660A1
公开(公告)日：2014-04-24
申请号：US14145318
申请日：2013-12-31
Applicant: Huawei Technologies Co., Ltd.
Inventor： Xiang Zhang , Hualin Wan , Jun Zhang
Abstract: A method and device for speaker recognition are provided. In the present invention, identifiability re-estimation is performed on a first vector (namely, a weight vector) in a score function by adopting a support vector machine (SVM), so that a recognition result of a characteristic parameter of a test voice is more accurate, thereby improving identifiability of speaker recognition.
Abstract translation: 提供了用于说话者识别的方法和装置。 在本发明中，通过采用支持向量机（SVM）对得分函数中的第一向量（即权重向量）进行可识别性再估计，使得测试语音的特征参数的识别结果为 更准确，从而提高说话人识别的识别能力。
-
公开(公告)号：US10511494B2
公开(公告)日：2019-12-17
申请号：US15644340
申请日：2017-07-07
Applicant: Huawei Technologies Co., Ltd.
Inventor： Hualin Wan , Xijun Xue , Tao Huang
IPC: H04L12/24 , G06Q20/06 , H04L29/06 , H04L12/851
Abstract: A network control method relates to the communications field, includes receiving, by a controller, a packet forwarded by a forwarder, detecting, by the controller, a status of a virtual currency identifier of the packet, querying, by the controller according to a user identifier in the packet, whether the user has permission to improve service quality when the status of the virtual currency identifier of the packet indicates that a user is willing to pay virtual currency to raise a network priority, and raising, by the controller, the network priority of the user, starting charging, and sending a network priority of the user to the forwarder when the user has the permission to improve the service quality such that the forwarder forwards a packet of the user according to the network priority of the user.
-
公开(公告)号：US09142210B2
公开(公告)日：2015-09-22
申请号：US14145318
申请日：2013-12-31
Applicant: Huawei Technologies Co., Ltd.
Inventor： Xiang Zhang , Hualin Wan , Jun Zhang , Li Lu
Abstract: A method and device for speaker recognition are provided. In the present invention, identifiability re-estimation is performed on a first vector (namely, a weight vector) in a score function by adopting a support vector machine (SVM), so that a recognition result of a characteristic parameter of a test voice is more accurate, thereby improving identifiability of speaker recognition.
Abstract translation: 提供了用于说话者识别的方法和装置。 在本发明中，通过采用支持向量机（SVM）对得分函数中的第一向量（即权重向量）进行可识别性再估计，使得测试语音的特征参数的识别结果为 更准确，从而提高说话人识别的识别能力。
-
公开(公告)号：US20170310561A1
公开(公告)日：2017-10-26
申请号：US15644340
申请日：2017-07-07
Applicant: Huawei Technologies Co., Ltd.
Inventor： Hualin Wan , Xijun Xue , Tao Huang
IPC: H04L12/24 , G06Q20/06 , H04L29/06 , H04L12/851
Abstract: A network control method relates to the communications field, includes receiving, by a controller, a packet forwarded by a forwarder, detecting, by the controller, a status of a virtual currency identifier of the packet, querying, by the controller according to a user identifier in the packet, whether the user has permission to improve service quality when the status of the virtual currency identifier of the packet indicates that a user is willing to pay virtual currency to raise a network priority, and raising, by the controller, the network priority of the user, starting charging, and sending a network priority of the user to the forwarder when the user has the permission to improve the service quality such that the forwarder forwards a packet of the user according to the network priority of the user.
-
-
-
Search Results
4
records
(took 0.014 seconds)
