Patents Examined by Marcus T. Riley
  • Patent number: 10796098
    Abstract: A new technology of prediction of manipulability in response even to an instruction with missing information in an object manipulation task to have a robot manipulate some kind of object is provided. An instruction understanding system includes an obtaining engine configured to obtain a linguistic expression of a name of an object to be manipulated and a linguistic expression of a situation where the object corresponding to the name is placed in a real environment and a classifier configured to receive input of the linguistic expression of the name and the linguistic expression of the situation and output manipulability of the object corresponding to the name in the real environment.
    Type: Grant
    Filed: September 6, 2018
    Date of Patent: October 6, 2020
    Assignee: NATIONAL INSTITUTE OF INFORMATION AND COMMUNICATIONS TECHNOLOGY
    Inventors: Komei Sugiura, Hisashi Kawai
  • Patent number: 10789954
    Abstract: According to one or more aspects of the present disclosure, operations related to providing transcriptions may include obtaining a first transcription of first audio obtained by a first device during a communication session conducted between the first device and a second device. The operations may further include providing the first transcription for presentation of the first transcription by a display device during the communication session. In addition, the operations may include providing, in response to a transcription quality indication, a second transcription of second audio obtained by the second device during the communication session for presentation of the second transcription by the display device during the communication session.
    Type: Grant
    Filed: August 29, 2018
    Date of Patent: September 29, 2020
    Assignee: Sorenson IP Holdings, LLC
    Inventor: Michael Holm
  • Patent number: 10783904
    Abstract: A method, and device, for enhancing speech generated from bone and tissue conduction of a user of an In-ear device in a noisy environment, the Intra-aural device having an in-ear microphone adapted to be in fluid communication with the ear canal of the user and an outer-ear microphone adapted to be in fluid communication with the environment outside the ear. The method comprises applying an adaptive filter on the in-ear microphone signal, using the outer-ear microphone signal as a reference for the ambient noise and interrupting the application of the adaptive filter to the In-ear microphone signal upon detecting speech by the user.
    Type: Grant
    Filed: May 10, 2017
    Date of Patent: September 22, 2020
    Assignee: EERS GLOBAL TECHNOLOGIES INC.
    Inventors: Rachel E. Bouserhal, Jérémie Voix, Tiago Falk
  • Patent number: 10783331
    Abstract: The present disclosure provides a method and apparatus for building a text classification model, and a text classification method and apparatus. The method of building a text classification model comprises: obtaining a training sample; obtaining a vector matrix corresponding to the text, after performing word segmentation for the text based on an entity dictionary; using the vector matrix corresponding to the text and a class of the text to train a first classification model and a second classification model respectively; during the training process, using a loss function of the first classification model and a loss function of the second classification model to obtain a loss function of the text classification model, and using the loss function of the text classification model to adjust parameters for the first classification model and the second classification model, to obtain the text classification model formed by the first classification model and the second classification model.
    Type: Grant
    Filed: August 29, 2018
    Date of Patent: September 22, 2020
    Assignee: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD.
    Inventors: Yuan Xia, Jingbo Zhou, Weishan Dong, Wei Fan
  • Patent number: 10768871
    Abstract: Output devices supporting wireless device discovery are herein disclosed and enabled. To setup an output device for wireless connection, wireless device discovery is performed by a client of the output device to discover the output device. Upon being discovered, the output device may provide identification or information related to the output device to the client for establishing a wireless connection. Subsequently, the output device may receive, via the established wireless connection, digital content from the client for output. Wireless device discovery operations may be based on one or more of a close proximity wireless communication, a Bluetooth wireless communication, or a wireless communication compatible with a protocol within IEEE 802.11 standards. Security information or authentication may be required for establishing the wireless connection. Subsequent to having established the wireless connection, the output device receives the digital content from the client for playing at the output device.
    Type: Grant
    Filed: July 31, 2018
    Date of Patent: September 8, 2020
    Assignee: Flexiworld Technologies, Inc.
    Inventors: William Ho Chang, Christina Y. Liu
  • Patent number: 10771408
    Abstract: A chatbot system and method conversing with other chatbots in place of a user. The chatbot system and method understand the purpose of conversation of a user, select chatbots to which the purpose of the user is to be transferred, converse with the selected chatbots in place of the user, and present the user with a result of conversation undertaken. The chatbot system converses with provider chatbots provided by property and service providers. The chatbot system includes a consumer chatbot that understands a purpose of conversation of a user, converses with the provider chatbots in place of the user, and provides a result of the conversation to the user.
    Type: Grant
    Filed: September 6, 2018
    Date of Patent: September 8, 2020
    Assignee: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE
    Inventors: Joon Ho Lim, Hyun Ki Kim, Min Ho Kim, Ji Hee Ryu, Kyung Man Bae, Yong Jin Bae, Ji Hyun Wang, Hyung Jik Lee, Soo Jong Lim, Myung Gil Jang, Mi Ran Choi, Jeong Heo
  • Patent number: 10769388
    Abstract: An approach is provided for changing a language for a replay of a session of a user. Request and response data in the session is captured. If a preferred language for the replay does not match an initial language of content presented to the user in the session, the captured data is translated into the preferred language and the content is presented in the preferred language and not in the initial language. If the preferred language matches the initial language, the content is presented in the initial language, without translating the captured data. An event in the captured data indicates a change in a displayed image. Responsive to searching Hypertext Markup Language (HTML) documents generated from a resource file, the HTML documents are determined to not include an HTML document that specifies the change. The HTML document is generated and employed in the replay of the session.
    Type: Grant
    Filed: November 25, 2019
    Date of Patent: September 8, 2020
    Assignee: International Business Machines Corporation
    Inventors: Sunil Chelani, Malarvizhi Kandasamy, Mali Hansraj, Mohammad N. Nazmi
  • Patent number: 10761791
    Abstract: Wireless printing devices that are configured to register with a service over a network (e.g., the Internet) for providing printing services without a device specific printer driver are herein disclosed and enabled. The printing device may include an interface (e.g., a touch sensitive screen) to receive security information for connecting to a wireless local area network (WLAN). While connected to the wireless WLAN, the printing device registers the printing device with the service and transmits, to the service, device information related to the printing device. Subsequent to registration, one or more client devices, using the service over the network, may select the registered printing device for printing. The printing device is configured to receive output data either from the service over the network or from the one or more client devices over the network. The output data is related to the device information transmitted from the printing device to the service.
    Type: Grant
    Filed: October 6, 2017
    Date of Patent: September 1, 2020
    Assignee: Flexiworld Technologies, Inc.
    Inventors: William Ho Chang, Christina Ying Liu
  • Patent number: 10755715
    Abstract: A restocking system includes a first compartment for storing consumable items; a restocking suggestion unit configured to receive restocking suggestions from a remotely located server; an output interface configured to output the restocking suggestions to a user; and a first hardware button located in an interior of the first compartment. The first hardware button is configured to: after a first restocking suggestion is output by the output interface, prompt the user to provide a confirmation input regarding the first restocking suggestion; send a respective confirmation signal to the remotely located server if the first hardware button is activated by the user within a preset time window after the first restocking suggestion is output; and send a respective rejection signal to the remotely located server if the first hardware button is not activated within the preset time window. A corresponding restocking method is also disclosed.
    Type: Grant
    Filed: February 24, 2020
    Date of Patent: August 25, 2020
    Assignee: MIDEA GROUP CO., LTD.
    Inventors: Yingbiao Huang, Tian Tang
  • Patent number: 10733389
    Abstract: A computer-implemented method for leveraging computer aided input segmentations is provided. The computer aided input segmentations identify divisions in a source text. The division are utilized to translate the source text from a first language to a second language. In this regard, segmentation boundaries are designated within the source text based on the computer aided input segmentations, which determine the divisions. Each division is automatically translated into the second language to generate translated segments corresponding to the division. These translated segments are combined to generate a translated text.
    Type: Grant
    Filed: September 5, 2018
    Date of Patent: August 4, 2020
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Fei Huang, Jian-Ming Xu
  • Patent number: 10733980
    Abstract: A recurrent neural network (RNN) is trained to identify split positions in long content, wherein each split position is a position at which the theme of the long content changes. Each sentence in the long content is converted to a vector that corresponds to the meaning of the sentence. The sentence vectors are used as inputs to the RNN. The high-probability split points determined by the RNN may be combined with contextual cues to determine the actual split point to use. The split points are used to generate thematic segments of the long content. The multiple thematic segments may be presented to a user along with a topic label for each thematic segment. Each topic label may be generated based on the words contained in the corresponding thematic segment.
    Type: Grant
    Filed: April 17, 2019
    Date of Patent: August 4, 2020
    Assignee: SAP SE
    Inventors: Jayananda Appanna Kotri, Tarun Sharma, Sharad Kejriwal, Yashwanth Dasari, Abinaya S
  • Patent number: 10726054
    Abstract: A system and method for generating at least one policy includes a policy document database containing at least one policy document containing at least one unstructured policy entry, and a natural language processor to analyze the at least one unstructured policy entry to generate least one formal policy, wherein a formal outcome of execution of the at least one formal policy corresponds to the at least one unstructured policy entry, and a rule processor to transform the at least one formal policy entry to generate at least one enforceable policy, wherein an enforcement outcome of execution of the at least one enforceable policy corresponds to the at least one formal policy entry.
    Type: Grant
    Filed: February 21, 2017
    Date of Patent: July 28, 2020
    Assignee: CARRIER CORPORATION
    Inventors: Blanca Florentino, Tarik Hadzic, Suresh B. Veluru, Ankit Tiwari
  • Patent number: 10726207
    Abstract: Methods, systems, and computer-readable storage media for receiving a set of document-level training data including a plurality of documents, each document having a sentiment label associated therewith, receiving a set of aspect-level training data including a plurality of aspects, each aspect having a sentiment label associated therewith, training the aspect-level sentiment classifier including a long short-term memory (LSTM) network, and an output layer using one or more of pretraining, and multi-task learning based on the document-level training data and the aspect-level training data, pretraining including initializing parameters based on pretrained weights that are fine-tuned during training, and multi-task learning including simultaneous training of document-level classification and aspect-level classification, and providing the aspect-level sentiment classifier for classifying one or more aspects in one or more sentences of one or more input documents based on sentiment classes.
    Type: Grant
    Filed: November 27, 2018
    Date of Patent: July 28, 2020
    Assignee: SAP SE
    Inventor: Ruidan He
  • Patent number: 10726826
    Abstract: A computer system receives a training set of voice data captured from one or more speakers and classified with one or more categorical prosodic labels according to one or more prosodic categories. The computer system transforms the voice data of the training set while preserving one or more portions of the voice data that determine the one or more categorical prosodic labels to automatically form a new training set of voice data automatically classified with the one or more categorical prosodic labels according to the one or more prosodic categories. The computer system augments a database comprising the training set with the new training set for training a speech model of an artificial intelligence system.
    Type: Grant
    Filed: March 4, 2018
    Date of Patent: July 28, 2020
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Raul Fernandez, Andrew Rosenberg, Alexander Sorin
  • Patent number: 10706210
    Abstract: In an automatic speech recognition (ASR) dictation application, a user interface may be provided for informing a user how to dictate desired text. Input may be received from the user of the dictation application, specifying a desired text sequence. In response to the received input, output may automatically be provided to the user via the user interface, indicating one or more speech sequences that, when spoken by a user and recognized by the dictation application using ASR, would cause the dictation application to output the desired text sequence as a recognition result.
    Type: Grant
    Filed: August 31, 2016
    Date of Patent: July 7, 2020
    Assignee: Nuance Communications, Inc.
    Inventor: Kaarel Kaljurand
  • Patent number: 10699060
    Abstract: A system includes a neural network for performing a first natural language processing task. The neural network includes a first rectifier linear unit capable of executing an activation function on a first input related to a first word sequence, and a second rectifier linear unit capable of executing an activation function on a second input related to a second word sequence. A first encoder is capable of receiving the result from the first rectifier linear unit and generating a first task specific representation relating to the first word sequence, and a second encoder is capable of receiving the result from the second rectifier linear unit and generating a second task specific representation relating to the second word sequence. A biattention mechanism is capable of computing, based on the first and second task specific representations, an interdependent representation related to the first and second word sequences.
    Type: Grant
    Filed: June 5, 2018
    Date of Patent: June 30, 2020
    Assignee: salesforce.com, inc.
    Inventors: Bryan McCann, Caiming Xiong, Richard Socher
  • Patent number: 10699695
    Abstract: During text-to-speech processing, audio data corresponding to a word part, word, or group of words is generated using a trained model and used by a unit selection engine to create output audio. The audio data is generated at least when an input word is unrecognized or when a cost of a unit selection is too high.
    Type: Grant
    Filed: June 29, 2018
    Date of Patent: June 30, 2020
    Assignee: Amazon Washington, Inc.
    Inventors: Adam Franciszek Nadolski, Daniel Korzekwa, Thomas Edward Merritt, Marco Nicolis, Bartosz Putrycz, Roberto Barra Chicote, Rafal Kuklinski, Wiktor Dolecki
  • Patent number: 10657981
    Abstract: Techniques for improving acoustic echo cancellation to attenuate an echo signal generated by a loudspeaker included in a device are described. A system may determine a loudspeaker canceling beam (LCB) (e.g., fixed beam directed to the loudspeaker) and may use the LCB to generate LCB audio data that corresponds to the echo signal. For example, based on a configuration of the loudspeaker relative to microphone(s) of the device, the system may perform simulation(s) to generate a plurality of filter coefficient values corresponding to the loudspeaker. By subtracting the LCB audio data during acoustic echo cancellation, the system may attenuate the echo signal even when there is distortion or nonlinearity or the like caused by the loudspeaker. In some examples, the system may perform acoustic echo cancellation using the LCB audio data and playback audio data.
    Type: Grant
    Filed: May 17, 2018
    Date of Patent: May 19, 2020
    Assignee: Amazon Technologies, Inc.
    Inventors: Mohamed Mansour, Robert Ayrapetian
  • Patent number: 10650836
    Abstract: Example embodiments disclosed herein relate to signal processing. A method for decomposing a plurality of audio signals from at least two different channels is disclosed. The method comprises obtaining a set of components that are weakly correlated, the set of components generated based on the plurality of audio signals. The method comprises extracting a feature from the set of components, and determining a set of gains associated with the set of components at least in part based on the extracted feature, each of the gains indicating a proportion of a diffuse part in the associated component. The method further comprises decomposing the plurality of audio signals by applying the set of gains to the set of components. Corresponding system and computer program product are also disclosed.
    Type: Grant
    Filed: September 20, 2019
    Date of Patent: May 12, 2020
    Assignee: Dolby Laboratories Licensing Corporation
    Inventors: Jun Wang, Lie Lu
  • Patent number: 10650818
    Abstract: Disclosed herein is a method for performing semi-supervised question answering. The method includes the steps of: (i) creating a graph based on input data, the input data comprising natural language text that includes seen questions, background text, and seen answers, and (ii) querying the graph in response to a new question. Querying the graph in response to a new question includes the steps of: projecting the question into the graph to a projected location; and for each node in a retrieval radius associated with the question, determining a set of answer weights based on the distances between retrieved nodes and the projected location and answer weights associated with retrieved nodes based on seen questions; and determining a best response corresponding to the seen answer associated with the highest answer weight from the set of answer weights determined for each retrieved node within the retrieval radius.
    Type: Grant
    Filed: August 31, 2018
    Date of Patent: May 12, 2020
    Assignee: FLAMINGO AI LIMITED
    Inventor: Jack Winston Elliott