Patents Examined by Stella L. Woo
  • Patent number: 12223270
    Abstract: Embodiments described herein provide a method of evaluating a natural language processing model. The method includes receiving an evaluation dataset that may include a plurality of unit tests, the unit tests having: an input context, and a first candidate and a second candidate that are generated in response to the input context, where the first test candidate is associated with a first quality notation, and the second candidate is associated with a second quality notation. The method includes determining, via a model, a first likelihood of generating the first candidate and a second likelihood of generating the second candidate in response to the input context. The method also includes determining whether the first likelihood being greater than the second likelihood. The method also includes determining whether the first model passed the unit test, where the first quality notation indicates a higher quality candidate and the second quality notation indicate a lower quality candidate.
    Type: Grant
    Filed: June 10, 2022
    Date of Patent: February 11, 2025
    Assignee: Salesforce, Inc.
    Inventors: Philippe Laban, Chien-Sheng Wu, Wenhao Liu, Caiming Xiong
  • Patent number: 12216962
    Abstract: Aspects of the present disclosure are directed to quantitatively tracking food intake using smart glasses and/or other wearable devices. In some implementations, the smart glasses can include an image capture device, such as a camera, that can seamlessly capture images of food being eaten by the user. A computing device in communication with the smart glasses (or the smart glasses themselves) can identify the type and volume of food being eaten by applying object recognition and volume estimation techniques to the images. Additionally or alternatively, the smart glasses and/or other wearable devices can track a user's eating patterns through the number of bites taken throughout the day by capturing and analyzing hand-to-mouth motions and chewing. The computing device can log the type of food, volume of food, and/or number of bites taken and compute statistics that can be displayed to the user on the smart glasses.
    Type: Grant
    Filed: July 19, 2022
    Date of Patent: February 4, 2025
    Assignee: Meta Platforms Technologies, LLC
    Inventors: Wen Li, John Rumsfeld, Charles Liam Goudge, Freddy Abnousi
  • Patent number: 12212727
    Abstract: Embodiments of a fax system with adaptive protocol selection, and methods for such a system, are disclosed herein. Embodiments of a fax system may be adapted to selectively configure the protocol (both the type of protocol or aspects of a particular protocol) used in association with the transmission or reception of a specific fax. The configuration of the protocol utilized can be based on one or more attributes associated with a sender or a destination. The fax can then be transmitted or received according to those configuration parameters.
    Type: Grant
    Filed: October 20, 2023
    Date of Patent: January 28, 2025
    Assignee: OPEN TEXT HOLDINGS, INC.
    Inventors: Michael Alexander Preobrazhensky, Philippe C. Haussy, David M. Hayden
  • Patent number: 12204850
    Abstract: Embodiments of the present disclosure relate to semantic parsing for short text. According to embodiments of the present disclosure, a method is proposed. The method comprises: obtaining a set of sentences associated with a short text, each of the set of sentences containing all of words in the short text; determining a set of syntactic features associated with the set of sentences, each of the set of syntactic features indicating at least one of a constituency relation and a dependency relation of the corresponding sentence; and determining a semantic structure of the short text based on the set of syntactic features.
    Type: Grant
    Filed: April 25, 2022
    Date of Patent: January 21, 2025
    Assignee: LEMON INC.
    Inventors: Hongyu Xiong, Han Wang, Rui Li, Yiqi Feng, Yuan Gao, Xiangyu Zeng, Qingyi Lu, Bin Liu
  • Patent number: 12205026
    Abstract: Methods and systems for language processing include augmenting an original training dataset to produce an augmented dataset that includes a first example that includes a first scrambled replacement for a first word and a definition of the first word, and a second example that includes a second scrambled replacement for the first word and a definition of an alternative to the first word. A neural network classifier is trained using the augmented dataset.
    Type: Grant
    Filed: February 17, 2022
    Date of Patent: January 21, 2025
    Assignee: NEC Corporation
    Inventor: Christopher Malon
  • Patent number: 12198694
    Abstract: An information processing apparatus according to the present disclosure includes an acquisition unit that acquires inspiration information indicating inspiration of a user, and a prediction unit that predicts whether or not the user utters after the inspiration of the user on the basis of the inspiration information acquired by the acquisition unit.
    Type: Grant
    Filed: January 12, 2021
    Date of Patent: January 14, 2025
    Assignee: SONY GROUP CORPORATION
    Inventors: Yuhei Taki, Yasuo Kabe, Hiro Iwase, Kunihito Sawai
  • Patent number: 12190068
    Abstract: Methods and systems are presented for generating real-time dynamic conversational responses during conversational interactions using machine learning models based on historic intents for a plurality of users and user-specific interactions. The machine learning models comprise a neural network trained to select a first intent from a plurality of intents based on historic data accumulated prior to the conversational interaction and a neural network trained to select a first interaction-specific intent from a plurality of interaction-specific intents based on interaction-specific data for a user.
    Type: Grant
    Filed: June 27, 2022
    Date of Patent: January 7, 2025
    Assignee: Capital One Services, LLC
    Inventors: Md Arafat Hossain Khan, Isha Chaturvedi, Arturo Hernandez Zeledon, Mohammad Sorower
  • Patent number: 12190907
    Abstract: There are provided a method and a system for automatically generating a back-channel in an interactive agent system. According to an embodiment of the disclosure, an automatic back-channel generation method includes: predicting a back-channel by analyzing an utterance of a user inputted in a back-channel prediction model; and generating the predicted back-channel, and the back-channel prediction model is an AI model that is trained to predict a back-channel to express from the utterance of the user. Accordingly, a back-channel is automatically generated by utilizing a back-channel prediction module which is based on a language model, so that a natural dialogue interaction with a user may be implemented in an interactive agent system, and quality of a dialogue service provided to a user may be enhanced.
    Type: Grant
    Filed: August 23, 2022
    Date of Patent: January 7, 2025
    Assignee: KOREA ELECTRONICS TECHNOLOGY INSTITUTE
    Inventors: Jin Yea Jang, Sa Im Shin, Min Young Jung, San Kim
  • Patent number: 12183337
    Abstract: Methods, systems, and apparatuses are described for context-based speech processing. A current context may be determined. A subset of trigger words or phrases may be selected, from a plurality of trigger words, as trigger words that a computing device is configured to recognize in the determined current context. The computing device may be controlled to listen for only the subset of trigger words during speech recognition.
    Type: Grant
    Filed: July 20, 2021
    Date of Patent: December 31, 2024
    Assignee: Comcast Cable Communications, LLC
    Inventor: Michael Chen
  • Patent number: 12182498
    Abstract: Portions of text data generated from inverse text normalization may be redacted. Text data for redaction may be obtained. One or more inverse text normalization models may be applied to the text data to generate normalized text data. A machine learning model, trained to recognize text for redaction, may be applied to identify portions of the normalized text data for redaction. The identified portions may be redacted and the redacted normalized text provided to a destination.
    Type: Grant
    Filed: June 30, 2022
    Date of Patent: December 31, 2024
    Assignee: Amazon Technologies, Inc.
    Inventors: Monica Lakshmi Sunkara, Deepthi Devaiah Devanira, Chaitanya Shivade, Sravan Babu Bodapati, Katrin Kirchhoff, Srikanth Ronanki
  • Patent number: 12175571
    Abstract: One variation of a method for securing synthetic video conference feeds includes, during a setup period: accessing a target image depicting a face of a user; generating a face model based on a target set of facial features detected in the target image; and linking the face model to a target set of facial biometric values of the user. The method also includes, during an operating period succeeding the setup period: accessing a frame; deriving characteristics of a set of facial features detected in the frame; extracting a set of facial biometric values from the frame; in response to alignment between the target set of facial biometric values and the set of facial biometric values, generating a synthetic face image based on characteristics of the set of facial features, the face model, and a synthetic face generator; and rendering the synthetic face image in place of the frame.
    Type: Grant
    Filed: May 4, 2022
    Date of Patent: December 24, 2024
    Assignee: Present Communications, Inc.
    Inventors: Yousif Astarabadi, Matt Mireles, Shaun Astarabadi, Kristina Nikkhah
  • Patent number: 12170699
    Abstract: A system and method are used to manage GUI for group video calls. The method ensures effective communication and control over shared video content during group video calls. The method enables to combine feed videos from a plurality of attendee accounts and other shared video content to be displayed in one single slide. Further, the method allows control over slide presentations and positioning of participant videos by cursor movement of the host account holder or the presenter. This allows the participants to follow the presenter, while viewing the presentation slides and other feed videos. Furthermore, the method allows users to switch mouse cursor control and feed video positions between the presenter and the participants by clicking on the video on a primary slot. Additionally, the host has special controls that may be passed to a selected number of participants and the participant feed videos are arranged in two sections.
    Type: Grant
    Filed: January 9, 2024
    Date of Patent: December 17, 2024
    Inventor: Juergen Koehler
  • Patent number: 12165648
    Abstract: An electronic device and method are disclosed. The electronic device includes input circuitry, a display, and a processor. The processor implements the method, including extracting at least one piece of context information based at least in part on an application screen displayed on the display, analyzing the extracted at least one piece of context information to generate a language model based on the extracted at least one piece of context information, receiving a voice input of a user through the input circuitry and convert the voice input into a text string using the generated language model, and resetting the generated language model.
    Type: Grant
    Filed: May 25, 2022
    Date of Patent: December 10, 2024
    Assignee: Samsung Electronics Co., Ltd.
    Inventors: Kihyung Nam, Seokyeong Jung, Jungkeun Cho, Jinwoong Kim
  • Patent number: 12135948
    Abstract: Bias scores can be generated from articles. A bias score generator can be configured to aggregate and process articles to identify entities and policies that appear in the articles and to generate sentiment values for the entities and policies. Aspect-based targeted sentiment analysis may be employed to generate the sentiment values. The bias score generator can process the sentiment values to generate a bias score for each article. Bias scores for multiple articles can be combined to generate a media bias score for a source of the articles. Bias loaded words and bias loaded snippets may also be used to generate a media bias score.
    Type: Grant
    Filed: December 30, 2021
    Date of Patent: November 5, 2024
    Assignee: Biasly, LLC
    Inventors: Daniel Carter, Haochuan Wei, Arjun Karangiya, William Lucia
  • Patent number: 12125479
    Abstract: A system for providing a sociolinguistic virtual assistant includes a communication device, a processing device, and a storage device. The processing device being configured to process input data using a natural language processing algorithm; categorize the semantic data based on psych-sociological categorizations associated with the at least one user; analyze the command from the at least one user to identify a task associated with the command; generate a response based on identification of the task associated with the command; execute the task associated with the command using categorized semantic data, to derive a result. A method corresponding to the system is also provided.
    Type: Grant
    Filed: February 8, 2022
    Date of Patent: October 22, 2024
    Assignee: Seam Social Labs Inc
    Inventors: Tiasia O'Brien, Marisa Jean Dinko
  • Patent number: 12108189
    Abstract: A communication server facilitates a video call between client devices of a plurality of participants. During the video call, the communication server identifies an in-call activity to recommend to the plurality of participants in the video call based on user information. The in-call activity is selected to be relevant to each of the plurality of participants and jointly recommended to the plurality of participants of the video call. To identify the recommended in-call activity, the communication server may determine common interests among the plurality of participants and select an in-call activity that is associated with the common interests. After the recommended in-call activity is selected, an indication of the recommended in-call activity is provided to the client devices to enable the client devices to display a user interface including the in-call activity to the participants during the video call.
    Type: Grant
    Filed: January 22, 2022
    Date of Patent: October 1, 2024
    Assignee: Meta Platforms, Inc.
    Inventors: Suchada Sutasirisap, John Kilcline, Tomas Brennessl, Tianyu Li
  • Patent number: 12086543
    Abstract: A system and method for creating a machine learning (ML) classifier for a database uses a weakly-supervised training data set created automatically from database items on the basis of a human-created keyword set. The automatically created training data set is used to construct one or more deep learning classifier checkpoints, which can then be compared with one another and with a classifier based on the original keyword set in order to select a classifier for use by other users viewing the database.
    Type: Grant
    Filed: June 23, 2021
    Date of Patent: September 10, 2024
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Sathia Prabhu Thirumal, Christopher Lawrence Laterza, Manoj Kumar Rawat, Karan Singh Rekhi, Natarajan Arumugam, Pranav Jayant Farswani
  • Patent number: 12086565
    Abstract: Mechanisms for implementing a text encoder and text encoder operations are provided. A contrastive machine learning training operation trains an encoder of a machine learning computer model, to learn a sense and similarity preserving embedding that operates to encode input natural language text data to generate encoded natural language text data based on a sense attribute of one or more terms in the input natural language text data. The contrastive machine learning training operation learns to separate positive samples in training data from negative samples in the training data. The trained encoder processes a term specified in an input natural language text to generate an encoded natural language text based on the learned embedding and inputs, to a downstream computing system, the encoded natural language text to cause the downstream computing system to perform a computer natural language processing operation based on the embedding.
    Type: Grant
    Filed: February 28, 2022
    Date of Patent: September 10, 2024
    Assignee: International Business Machines Corporation
    Inventor: Tanveer Syeda-Mahmood
  • Patent number: 12079572
    Abstract: A system and method for creating a machine learning (ML) classifier for a database uses a weakly-supervised training data set created automatically from database items on the basis of a human-created keyword set. The automatically created training data set is used to construct one or more deep learning classifier checkpoints, which can then be compared with one another and with a classifier based on the original keyword set in order to select a classifier for use by other users viewing the database.
    Type: Grant
    Filed: May 17, 2021
    Date of Patent: September 3, 2024
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Sathia Prabhu Thirumal, Christopher Lawrence Laterza, Manoj Kumar Rawat, Karan Singh Rekhi, Natarajan Arumugam, Pranav Jayant Farswani
  • Patent number: 12062357
    Abstract: A method of registering an attribute in a speech synthesis model, an apparatus of registering an attribute in a speech synthesis model, an electronic device, and a medium are provided, which relate to a field of an artificial intelligence technology such as a deep learning and intelligent speech technology. The method includes: acquiring a plurality of data associated with an attribute to be registered; and registering the attribute in the speech synthesis model by using the plurality of data associated with the attribute, wherein the speech synthesis model is trained in advance by using a training data in a training data set.
    Type: Grant
    Filed: November 16, 2021
    Date of Patent: August 13, 2024
    Assignee: Beijing Baidu Netcom Science Technology Co., Ltd.
    Inventors: Wenfu Wang, Xilei Wang, Tao Sun, Han Yuan, Zhengkun Gao, Lei Jia