Patents by Inventor Guoguo Chen

Guoguo Chen has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 11322153
    Abstract: A conversation interaction method and apparatus, and a computer-readable storage medium are provided. The method includes: converting a speech to be recognized into a first text; inputting the first text into a semantic analysis model, to obtain intention information and slot information of the first text; and inputting the intention information and the slot information of the first text into a conversation state machine, to obtain interaction information corresponding to the first text. By using a semantic analysis model, intention information and slot information of a first text are obtained directly from the first text. The process in the existing technology, where a semantic analysis model needs to be used immediately after a language model, is avoided, thereby shortening processing time and making it possible to respond faster to a user. Further, by using the above scheme, calculation complexity and the cost of a whole system are reduced.
    Type: Grant
    Filed: February 21, 2020
    Date of Patent: May 3, 2022
    Assignees: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD., SHANGHAI XIAODU TECHNOLOGY CO. LTD.
    Inventors: Yunfei Xu, Guoguo Chen
  • Patent number: 11212628
    Abstract: The present disclosure discloses a method and an apparatus for testing a speaker, an electronic device and a storage medium. A specific implementation includes: obtaining first audio data recorded by a microphone integrated with the speaker in ambient white noise; analyzing the first audio data to derive a first analysis result; and determining whether there is a defect in the microphone according to the first analysis result. Hence, these allow for testing a completed set on an assembled speaker to ensure the consistency of a microphone test and improve the accuracy of the test result.
    Type: Grant
    Filed: March 17, 2020
    Date of Patent: December 28, 2021
    Assignees: Baidu Online Network Technology (Beijing) Co., Ltd., Shanghai Xiaodu Technology Co., Ltd.
    Inventors: Aihui An, Ming Yu, Kang Lei, Guoguo Chen
  • Patent number: 10979835
    Abstract: The present disclosure discloses a method and an apparatus for testing a speaker, an electronic device and a storage medium. A specific implementation includes: obtaining first audio data recorded by a microphone integrated with the speaker in ambient white noise; analyzing the first audio data to derive a first analysis result; and determining whether there is a defect in the microphone according to the first analysis result. Hence, these allow for testing a completed set on an assembled speaker to ensure the consistency of a microphone test and improve the accuracy of the test result.
    Type: Grant
    Filed: March 17, 2020
    Date of Patent: April 13, 2021
    Inventors: Aihui An, Ming Yu, Kang Lei, Guoguo Chen
  • Publication number: 20210058724
    Abstract: The present disclosure discloses a method and an apparatus for testing a speaker, an electronic device and a storage medium. A specific implementation includes: obtaining first audio data recorded by a microphone integrated with the speaker in ambient white noise; analyzing the first audio data to derive a first analysis result; and determining whether there is a defect in the microphone according to the first analysis result. Hence, these allow for testing a completed set on an assembled speaker to ensure the consistency of a microphone test and improve the accuracy of the test result.
    Type: Application
    Filed: March 17, 2020
    Publication date: February 25, 2021
    Applicant: Baidu Online Network Technology (Beijing) Co., Ltd.
    Inventors: Aihui AN, Ming YU, Kang LEI, Guoguo CHEN
  • Publication number: 20210043099
    Abstract: An apparatus, system, and method is disclosed for a hybrid approach to using AI agents and human agents to provide behavioral coaching. Hybrid modes of coaching are supported in which conversations can be handed off from AI agents to human agents. In some implementations, collaborate modes of coaching are supported in which a human agent collaborates with an AI agent.
    Type: Application
    Filed: August 6, 2020
    Publication date: February 11, 2021
    Inventors: Shenggang Du, Guoguo Chen
  • Publication number: 20210027788
    Abstract: A conversation interaction method and apparatus, and a computer-readable storage medium are provided. The method includes: converting a speech to be recognized into a first text; inputting the first text into a semantic analysis model, to obtain intention information and slot information of the first text; and inputting the intention information and the slot information of the first text into a conversation state machine, to obtain interaction information corresponding to the first text. By using a semantic analysis model, intention information and slot information of a first text are obtained directly from the first text. The process in the existing technology, where a semantic analysis model needs to be used immediately after a language model, is avoided, thereby shortening processing time and making it possible to respond faster to a user. Further, by using the above scheme, calculation complexity and the cost of a whole system are reduced.
    Type: Application
    Filed: February 21, 2020
    Publication date: January 28, 2021
    Inventors: Yunfei Xu, Guoguo Chen
  • Publication number: 20200213838
    Abstract: Embodiments of the present disclosure provide a method and apparatus for communication authentication processing, and an electronic device, where the method includes: transmitting, by a first device, a pairing request to a second device; receiving, by the first device, a pairing response transmitted by the second device, where the pairing response includes a first random value and first signature information, the first random value is configured to generate the first signature information; and acquiring, by the first device, a second random value and second signature information from a server according to the first random value and the first signature information, where the second random value and the second signature information are generated by the server according to the first random value and the first signature information, the second random value is configured to generate the second signature information.
    Type: Application
    Filed: December 19, 2019
    Publication date: July 2, 2020
    Applicant: Baidu Online Network Technology (Beijing) Co., Ltd.
    Inventors: Peng WANG, Guoguo CHEN, Fei NIU, Ziqiang ZHU, Yin LONG
  • Publication number: 20200213844
    Abstract: Embodiments of the present disclosure provide a communication method, a communication apparatus and an electronic device. The method includes: a first device establishes a Bluetooth connection with a second device; the first device performs an authentication processing with the second device to obtain an authentication result; and the first device performs a voice-based data interaction with the second device if the authentication result is that an authentication is successful. The method can improve the connection speed, connection success rate and communication security during the communication of the device.
    Type: Application
    Filed: December 24, 2019
    Publication date: July 2, 2020
    Inventors: Peng WANG, Guoguo CHEN, Fei NIU, Aihui AN, Junlian HU
  • Patent number: 10666583
    Abstract: Embodiments of the inventive system and methods are directed to a computer program that employs a drag-and-drop user interface for managing dialogue states, tracking dialogue context, understanding dialogue utterances, and managing dialogue sessions. Each dialogue element is defined in a “node” that can be dragged and dropped into a canvas of the user interface. An embodiment provides wiring mechanisms to freely link nodes. Dialogue utterances are contained in messages that flow through the wires linking different nodes until exiting the canvas to an end user. An executable image of a conversational agent is then generated by compiling the source code associated with the nodes based on their connections. A conversational agent can be deployed in an electronic device such as a home device, which configured to perform an action in response to a user verbal command or request using the conversational agent deployed therein.
    Type: Grant
    Filed: November 27, 2017
    Date of Patent: May 26, 2020
    Assignee: BAIDU USA LLC
    Inventors: Xuchen Yao, Guoguo Chen
  • Patent number: 10600415
    Abstract: This disclosure provides a method, apparatus, device, and storage medium for voice interaction, where the method is applied to an AI device to determine whether a current scenario of the AI device is a preset scenario and waken a voice interaction function of the AI device to facilitate voice interaction with a user in response to the current scenario of the AI device being the preset scenario. A scenario directly triggers the voice interaction process, thereby avoiding the process of wakening by physical wakening or a wakening word, simplifying the process of using voice interaction, reducing the costs of learning voice interaction, and improving user experience.
    Type: Grant
    Filed: September 17, 2018
    Date of Patent: March 24, 2020
    Assignee: Baidu Online Network Technology (Beijing) Co., Ltd.
    Inventors: Jianan Xu, Guoguo Chen, Qinggeng Qian
  • Patent number: 10482903
    Abstract: A method for selectively interacting with multi-devices is provided. The method includes the following steps: receiving identical voice information transmitted by a plurality of terminal devices respectively; performing voice recognition on the received voice information; calculating energy of a wake-up word in respective voice information; and comparing the energy of one wake-up word with another, and transmitting feedback information to the terminal devices according to an energy comparison result and a voice recognition result. By calculating the energy of the wake-up word in respective voice information transmitted by respective devices, the distances between respective device and a user can be distinguished. A unique response can be ensured by determining that the device closest to the user responds to the user's request, thus ensuring the user experience.
    Type: Grant
    Filed: December 26, 2017
    Date of Patent: November 19, 2019
    Assignee: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD.
    Inventors: Sha Tao, Yonghui Zuo, Peng Wang, Guoguo Chen, Ji Zhou, Kaihua Zhu
  • Publication number: 20190198019
    Abstract: This disclosure provides a method, apparatus, device, and storage medium for voice interaction, where the method is applied to an AI device to determine whether a current scenario of the AI device is a preset scenario and waken a voice interaction function of the AI device to facilitate voice interaction with a user in response to the current scenario of the AI device being the preset scenario. A scenario directly triggers the voice interaction process, thereby avoiding the process of wakening by physical wakening or a wakening word, simplifying the process of using voice interaction, reducing the costs of learning voice interaction, and improving user experience.
    Type: Application
    Filed: September 17, 2018
    Publication date: June 27, 2019
    Inventors: Jianan XU, Guoguo CHEN, Qinggeng QIAN
  • Publication number: 20190166069
    Abstract: Embodiments of the inventive system and methods are directed to a computer program that employs a drag-and-drop user interface for managing dialogue states, tracking dialogue context, understanding dialogue utterances, and managing dialogue sessions. Each dialogue element is defined in a “node” that can be dragged and dropped into a canvas of the user interface. An embodiment provides wiring mechanisms to freely link nodes. Dialogue utterances are contained in messages that flow through the wires linking different nodes until exiting the canvas to an end user. An executable image of a conversational agent is then generated by compiling the source code associated with the nodes based on their connections. A conversational agent can be deployed in an electronic device such as a home device, which configured to perform an action in response to a user verbal command or request using the conversational agent deployed therein.
    Type: Application
    Filed: November 27, 2017
    Publication date: May 30, 2019
    Inventors: Xuchen Yao, Guoguo Chen
  • Publication number: 20190147904
    Abstract: A method for selectively interacting with multi-devices is provided. The method includes the following steps: receiving identical voice information transmitted by a plurality of terminal devices respectively; performing voice recognition on the received voice information; calculating energy of a wake-up word in respective voice information; and comparing the energy of one wake-up word with another, and transmitting feedback information to the terminal devices according to an energy comparison result and a voice recognition result. By calculating the energy of the wake-up word in respective voice information transmitted by respective devices, the distances between respective device and a user can be distinguished. A unique response can be ensured by determining that the device closest to the user responds to the user's request, thus ensuring the user experience.
    Type: Application
    Filed: December 26, 2017
    Publication date: May 16, 2019
    Inventors: Sha Tao, Yonghui Zuo, Peng Wang, Guoguo Chen, Ji Zhou, Kaihua Zhu
  • Patent number: 9754584
    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for recognizing keywords using a long short term memory neural network. One of the methods includes receiving, by a device for each of multiple variable length enrollment audio signals, a respective plurality of enrollment feature vectors that represent features of the respective variable length enrollment audio signal, processing each of the plurality of enrollment feature vectors using a long short term memory (LSTM) neural network to generate a respective enrollment LSTM output vector for each enrollment feature vector, and generating, for the respective variable length enrollment audio signal, a template fixed length representation for use in determining whether another audio signal encodes another spoken utterance of the enrollment phrase by combining at most a quantity k of the enrollment LSTM output vectors for the enrollment audio signal.
    Type: Grant
    Filed: November 8, 2016
    Date of Patent: September 5, 2017
    Assignee: Google Inc.
    Inventors: Maria Carolina Parada San Martin, Tara N. Sainath, Guoguo Chen
  • Patent number: 9715660
    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a deep neural network. One of the methods includes training a deep neural network with a first training set by adjusting values for each of a plurality of weights included in the neural network, and training the deep neural network to determine a probability that data received by the deep neural network has features similar to key features of one or more keywords or key phrases, the training comprising providing the deep neural network with a second training set and adjusting the values for a first subset of the plurality of weights, wherein the second training set includes data representing the key features of the one or more keywords or key phrases.
    Type: Grant
    Filed: March 31, 2014
    Date of Patent: July 25, 2017
    Assignee: Google Inc.
    Inventors: Maria Carolina Parada San Martin, Guoguo Chen, Georg Heigold
  • Publication number: 20170076717
    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for recognizing keywords using a long short term memory neural network. One of the methods includes receiving, by a device for each of multiple variable length enrollment audio signals, a respective plurality of enrollment feature vectors that represent features of the respective variable length enrollment audio signal, processing each of the plurality of enrollment feature vectors using a long short term memory (LSTM) neural network to generate a respective enrollment LSTM output vector for each enrollment feature vector, and generating, for the respective variable length enrollment audio signal, a template fixed length representation for use in determining whether another audio signal encodes another spoken utterance of the enrollment phrase by combining at most a quantity k of the enrollment LSTM output vectors for the enrollment audio signal.
    Type: Application
    Filed: November 8, 2016
    Publication date: March 16, 2017
    Inventors: Maria Carolina Parada San Martin, Tara N. Sainath, Guoguo Chen
  • Patent number: 9508340
    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for recognizing keywords using a long short term memory neural network. One of the methods includes receiving, by a device for each of multiple variable length enrollment audio signals, a respective plurality of enrollment feature vectors that represent features of the respective variable length enrollment audio signal, processing each of the plurality of enrollment feature vectors using a long short term memory (LSTM) neural network to generate a respective enrollment LSTM output vector for each enrollment feature vector, and generating, for the respective variable length enrollment audio signal, a template fixed length representation for use in determining whether another audio signal encodes another spoken utterance of the enrollment phrase by combining at most a quantity k of the enrollment LSTM output vectors for the enrollment audio signal.
    Type: Grant
    Filed: December 22, 2014
    Date of Patent: November 29, 2016
    Assignee: Google Inc.
    Inventors: Maria Carolina Parada San Martin, Tara N. Sainath, Guoguo Chen
  • Publication number: 20160180838
    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for recognizing keywords using a long short term memory neural network. One of the methods includes receiving, by a device for each of multiple variable length enrollment audio signals, a respective plurality of enrollment feature vectors that represent features of the respective variable length enrollment audio signal, processing each of the plurality of enrollment feature vectors using a long short term memory (LSTM) neural network to generate a respective enrollment LSTM output vector for each enrollment feature vector, and generating, for the respective variable length enrollment audio signal, a template fixed length representation for use in determining whether another audio signal encodes another spoken utterance of the enrollment phrase by combining at most a quantity k of the enrollment LSTM output vectors for the enrollment audio signal.
    Type: Application
    Filed: December 22, 2014
    Publication date: June 23, 2016
    Inventors: Maria Carolina Parada San Martin, Tara N. Sainath, Guoguo Chen
  • Patent number: D892711
    Type: Grant
    Filed: January 4, 2019
    Date of Patent: August 11, 2020
    Assignee: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD.
    Inventors: Tao Xiong, Yufeng Wang, Yuan Tian, Yaqian Zhang, Qinggeng Qian, Jingya Wang, Guoguo Chen