Tangome, Inc.

Chinese Academy of Sciences

Soundai Technology Co., Ltd.

Shaowei Su

Xu Liu

Yuanyuan Wang

Yixin Yang

Yuxin Liu

Yevgeni Litvin

Haim Vaisburd

Olivier Furon

Alexander Subbotin

Jinlin Wang

Le Chang

Hao Ren

Zhentan Feng

Dahang Feng

Xiaoliang Chen

Yifeng Lu

A method and device for fusing voiceprint features. The method includes: obtaining at least two voiceprint features of a voice sample of a target speaker (S; S); fusing the at least two voiceprint features on the basis of linear discriminant analysis (S). The present method introduces a technique employing linear discriminant analysis to fuse various voiceprint features, so as to improve complementarities between the various voiceprint features and distinctions between the fused features, thereby increasing the recognition rate for target speakers and reducing the misrecognition rate for non-target speakers in voiceprint authentication scenarios, and providing personalized and improved user experience.

Method and device for fusing voiceprint features, voice recognition method and system, and storage medium

Methods and systems for communicating with rate control. A communication is sent and received from a first device to a second device over a network, wherein the communication comprises at least one audio stream and a second communication stream. A capacity of the network is probed at the first device for the sending and receiving the communication. A presence of a voice in the at least one audio stream is detected at the first device via a voice activity detection of the at least one audio stream. A rate limit is set for the sending and receiving the communication at the first device based on the capacity of the network and the detection of the presence of the at least one audio stream.

Rate control for a communication

A method for controlling error resilience in network communication is described. The method includes: determining, by a receiver-side controller, a packet gap representing a packet loss of a packet being communicated over a network; projecting, by the receiver-side controller, a retransmission time-out for at least one missing packet of the packet loss; issuing, by the receiver-side controller, a retransmission request for the at least one missing packet; if the packet gap is not filled within a first time period of the retransmission time-out, then issuing, by the receiver-side controller, at least one synchronization frame request; and selecting, by a sender-side controller, to respond to at least one of either of the retransmission request or the at least one synchronization frame request and neither of the retransmission request nor the at least one synchronization frame request.

Joint retransmission and frame synchronization for error resilience control

The present invention relates to a method based on the combination of UPnP and STUN for NAT traversal. The method of the present invention effectively takes the advantages of UPnP and STUN. In particularly, when UpnP fails to traverse NAT, STUN is applied to perform the NAT traversal for the second time, so that the efficiency of the successful traversal is effectively improved. For the symmetric NAT traversal, a Port-Adding Sample Estimation method based on NAT adding port mapping is provided for port estimation, to avoid the defects resulted from the third-party agent and the blindness of the speculation approach. And, where both of the communication peers in the P2P streaming media system are behind NAT, the method that the part of the ordinary outer peers with the sufficient bandwidth acts as the Auxiliary-Relay peers is provided to release the pressure of the relay server so that the bottle-neck of the communication is broken through.

Method based on the combination of the UPnP and STUN for NAT traversal

Growing community of inventors

Beijing, China

Shaowei Su