-
1.
公开(公告)号:US20240194212A1
公开(公告)日:2024-06-13
申请号:US18553221
申请日:2022-03-25
Applicant: JINGDONG TECHNOLOGY HOLDING CO., LTD.
Inventor: Zehua CHEN , Junyi WU , Yuyu CAI , Wei XUE , Fan YANG , Guohong DING , Xiaodong HE
IPC: G10L21/0224 , G10L25/30 , G10L25/45
CPC classification number: G10L21/0224 , G10L25/30 , G10L25/45
Abstract: A method for end-to-end speech enhancement based on a neural network, including: obtaining a time-domain smoothing feature of an original speech signal by performing feature extraction on the original speech signal using a time-domain convolution kernel; and, obtaining an enhanced speech signal by performing combined feature extraction on the original speech signal and the time-domain smoothing feature of the original speech signal.
-
公开(公告)号:US20240119694A1
公开(公告)日:2024-04-11
申请号:US18257570
申请日:2022-04-08
Applicant: JINGDONG TECHNOLOGY HOLDING CO., LTD.
Inventor: Yongming HOU , Guangzhen SUI
CPC classification number: G06V10/22 , G06T11/001 , G06V40/16
Abstract: This disclosure provides an image data distribution method and apparatus, a terminal and a storage medium. The method includes generating a surface texture instance for processing information acquired by an image acquisition means; generating an image acquisition means instance corresponding to the image acquisition means; binding the image acquisition means instance with the surface texture instance; processing the information acquired by the acquisition means, generating image data by the surface texture instance; and acquiring the image data in an interface callback manner by using the image acquisition means instance, and distributing the image data to an application and/or a page instance.
-
公开(公告)号:US20230196351A1
公开(公告)日:2023-06-22
申请号:US17926354
申请日:2021-04-27
Applicant: Jingdong Technology Information Technology Co., Ltd. , Jingdong Technology Holding Co., Ltd.
Inventor: Xixia Gou
CPC classification number: G06Q20/389 , G06Q20/401 , H04L9/50 , H04L9/065
Abstract: A method and an apparatus for tracing a transaction based on a blockchain are provided. The method may include: invoking a smart contract between an upstream node and an intermediate node; encrypting, by using a private key of the intermediate node, transaction details between the upstream node and the intermediate node to generate and write upstream transaction information into the blockchain; encrypting the private key of the intermediate node by respectively using public keys of the upstream node and an end node, and generating and writing secret-key information of the upstream node and the end node into the blockchain; and decrypting the secret-key information of the upstream node and the end node by using the private keys of the upstream node and the end node, and decrypting the upstream transaction information of the intermediate node by using the private key of the intermediate node to obtain the transaction details.
-
公开(公告)号:US20230076658A1
公开(公告)日:2023-03-09
申请号:US17798298
申请日:2020-05-18
Applicant: JINGDONG TECHNOLOGY HOLDING CO., LTD.
IPC: G10L15/26
Abstract: Disclosed are a method, an apparatus, a computer device and a storage medium for decoding speech data. The method comprises acquiring at least one transcribed text obtained by transcribing the speech data; acquiring score of each transcribed text; acquiring at least one preset hot word corresponding to the speech data, each preset hot word corresponds to a reward value; and calculating, when there is a string matched with the preset hot word in the transcribed text, a target score of the transcribed text according to the reward value of the matched string and the score of the transcribed text, where the target score is used to determine the decoded text of the speech data. Hot word matching is performed on the transcribed text. If there is a matching hot word, the score of the transcribed text will be increased. The accuracy of decoding is improved without updating the model, and the operation is simple.
-
公开(公告)号:US20230039066A1
公开(公告)日:2023-02-09
申请号:US17784273
申请日:2020-05-15
Applicant: JINGDONG TECHNOLOGY INFORMATION TECHNOLOGY CO., LTD. , JINGDONG TECHNOLOGY HOLDING CO., LTD.
Abstract: Disclosed are a method and a system for cross-chain data access of blockchain. According to the technical solution, an adapter is added between a service blockchain and a witness blockchain, a cross-chain initiation event and a cross-chain response event on a local service blockchain are registered to the witness blockchain by using the adapter, so as to provide query and push services of cross-chain events to other nodes on the local service blockchain, and trigger related operations on the local service blockchain according to the cross-chain events of the witness blockchain.
-
公开(公告)号:US12182183B2
公开(公告)日:2024-12-31
申请号:US17785369
申请日:2020-04-20
Applicant: JINGDONG TECHNOLOGY HOLDING CO., LTD.
Inventor: Yuyu Zheng
IPC: G06F16/35 , G06F16/33 , G06F16/332 , H04L51/02
Abstract: The present application provides a robot response method, apparatus, device and storage medium. The method includes: obtaining, by a robot, current query voice; extracting semantic information of the current query voice; matching the semantic information of the current query voice with multiple semantic information clusters stored in advance to get a matched target semantic information cluster, where each semantic information cluster includes: at least one Q&A instance, and each Q&A instance includes: semantic information corresponding to a historical query voice and a query question selected in a query list corresponding to the historical query voice; and obtaining, by the robot, the number of times each query question was selected in the target semantic information cluster, determining, according to the number of times each query question was selected, a target query question corresponding to the current query voice, and outputting a query response corresponding to the target query question.
-
公开(公告)号:US12126732B2
公开(公告)日:2024-10-22
申请号:US17616543
申请日:2020-03-03
Applicant: JINGDONG TECHNOLOGY HOLDING CO., LTD.
Inventor: Zhuguang Shao , Haibo Sun , Guangwei Zhao , Haiquan Huang , Lin Zhang
CPC classification number: H04L9/3247 , H04L9/50
Abstract: The invention, which discloses a blockchain consensus method, apparatus and system, relates to the technical field of computers. A specific implementation mode of the method comprises: a leader node in the same network as a client receiving a transaction message sent by the client and a transaction message sent by at least one peer node, or receiving transaction messages sent by at least two peer nodes; sending the transaction messages to a leader node in a previous-level network till the priority of the previous-level network is the highest if transaction data in the transaction messages are consistent; a leader node in a highest-priority network generating a proposal message according to the received transaction messages, and performing a consensus judgment on the proposal message based on a Byzantine fault-tolerant consensus mechanism; sending the proposal message to a leader node in a next-level network till the next-level network is the network where the client is located after the success in the consensus judgment. The implementation mode can solve the technical problems of a long transaction confirmation time and a small transaction volume carried by a block.
-
公开(公告)号:US12094469B2
公开(公告)日:2024-09-17
申请号:US17603690
申请日:2020-03-03
Applicant: JINGDONG TECHNOLOGY HOLDING CO., LTD.
Inventor: Li Fu , Xiaoxiao Li
CPC classification number: G10L15/26 , G06F17/14 , G10L15/063 , G10L25/30
Abstract: A method for recognizing speech comprises: respectively setting initial values of a Chinese character coefficient and a Pinyin coefficient, generating a Chinese character mapping function according to the initial value of the Chinese character coefficient, and generating a Pinyin mapping function according to the initial value of the Pinyin coefficient (S101); training the Chinese character mapping function and the Pinyin mapping function using a plurality of preset training samples, calculating training results as parameters of a joint loss function, and generating a target mapping function according to calculation results (S102); and recognizing, according to the target mapping function, speech to be recognized, so as to obtain a Chinese character recognition result and a Pinyin recognition result of the speech to be recognized (S103). The method reduces the cost of speech recognition while ensuring the accuracy of speech recognition.
-
公开(公告)号:US20240155350A1
公开(公告)日:2024-05-09
申请号:US18550476
申请日:2021-12-23
Inventor: Zuoyi ZHANG
Abstract: The present disclosure provides a method and an apparatus for processing location-based service data, an electronic device, and a computer readable storage medium. The method includes: receiving a location service request sent by an IoT terminal, the location service request including a real address of the IoT terminal, location information, a type of the location service request, and a location service provider; replacing the real address of the IoT terminal in the location service request with a temporary address using a smart contract, and sending a replaced location service request to the location service provider; receiving a location service response sent by the location service provider; and replacing the temporary address included in the location service response with the real address using the smart contract, and sending a replaced location service response to the IoT terminal.
-
10.
公开(公告)号:US11979361B2
公开(公告)日:2024-05-07
申请号:US17753802
申请日:2020-04-13
Applicant: JINGDONG TECHNOLOGY HOLDING CO., LTD.
Inventor: Yuyu Zheng , Songxiang Gu , Jun Wang , Yu Zhang
IPC: H04L51/02 , G06F16/332
CPC classification number: H04L51/02 , G06F16/3329
Abstract: A dialogue robot generation method, a dialogue robot management platform, and a storage medium. The dialogue robot generation method comprises: obtaining at least one first function module associated with a dialogue robot to be generated (101); obtaining a calling sequence of the at least one first function module (102); and calling the at least one first function module on the basis of the calling sequence by means of a preset model container to generate a dialogue robot (103).
-
-
-
-
-
-
-
-
-