Patents by Inventor Ankur Gandhe
Ankur Gandhe 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: 11705116Abstract: Systems and methods described herein relate to adapting a language model for automatic speech recognition (ASR) for a new set of words. Instead of retraining the ASR models, language models and grammar models, the system only modifies one grammar model and ensures its compatibility with the existing models in the ASR system.Type: GrantFiled: August 18, 2021Date of Patent: July 18, 2023Assignee: Amazon Technologies, Inc.Inventors: Ankur Gandhe, Ariya Rastrow, Gautam Tiwari, Ashish Vishwanath Shenoy, Chun Chen
-
Publication number: 20220358908Abstract: Exemplary embodiments relate to adapting a generic language model during runtime using domain-specific language model data. The system performs an audio frame-level analysis, to determine if the utterance corresponds to a particular domain and whether the ASR hypothesis needs to be rescored. The system processes, using a trained classifier, the ASR hypothesis (a partial hypothesis) generated for the audio data processed so far. The system determines whether to rescore the hypothesis after every few audio frames (representing a word in the utterance) are processed by the speech recognition system.Type: ApplicationFiled: March 28, 2022Publication date: November 10, 2022Inventors: Ankur Gandhe, Ariya Rastrow, Roland Maximilian Rolf Maas, Bjorn Hoffmeister
-
Patent number: 11437043Abstract: Systems and methods for presence ground truth approximation and utilization are disclosed. For example, a system detects the presence of a predefined subject, such as a person associated with a given user profile, and/or determines that authentication criteria for performing an action in association with the user profile has been satisfied. A period of time to associate data is determined, and data of one or more data types is labeled as being associated with the speaker identification event. That data may be formatted and input into one or more models to train those models to more accurately detect presence and/or determine whether authentication of a user profile should succeed.Type: GrantFiled: December 12, 2019Date of Patent: September 6, 2022Assignee: Amazon Technologies, Inc.Inventors: Lizhen Peng, Alok Upadhyay, Jason Cline, Ankur Gandhe
-
Patent number: 11302310Abstract: Exemplary embodiments relate to adapting a generic language model during runtime using domain-specific language model data. The system performs an audio frame-level analysis, to determine if the utterance corresponds to a particular domain and whether the ASR hypothesis needs to be rescored. The system processes, using a trained classifier, the ASR hypothesis (a partial hypothesis) generated for the audio data processed so far. The system determines whether to rescore the hypothesis after every few audio frames (representing a word in the utterance) are processed by the speech recognition system.Type: GrantFiled: May 30, 2019Date of Patent: April 12, 2022Assignee: Amazon Technologies, Inc.Inventors: Ankur Gandhe, Ariya Rastrow, Roland Maximilian Rolf Maas, Bjorn Hoffmeister
-
Publication number: 20220036893Abstract: Systems and methods described herein relate to adapting a language model for automatic speech recognition (ASR) for a new set of words. Instead of retraining the ASR models, language models and grammar models, the system only modifies one grammar model and ensures its compatibility with the existing models in the ASR system.Type: ApplicationFiled: August 18, 2021Publication date: February 3, 2022Inventors: Ankur Gandhe, Ariya Rastrow, Gautam Tiwari, Ashish Vishwanath Shenoy, Chun Chen
-
Patent number: 11211058Abstract: Described herein is a system for prompting a user for clarification when an automatic speech recognition (ASR) system encounters ambiguity with respect to the user's input. The feedback provided by the user is used to retrain machine-learning models and/or to generate new machine-learning models. Based on the type of ambiguity, the system may determine to retrain one or more ASR models that are widely used by the system or to generate/update one or more user-specific models that are used to process inputs from one or more particular users.Type: GrantFiled: September 20, 2019Date of Patent: December 28, 2021Assignee: Amazon Technologies, Inc.Inventors: Aaron Eakin, Angela Sun, Ankur Gandhe, Ariya Rastrow, Chenlei Guo, Xing Fan
-
Patent number: 11145296Abstract: Systems and methods described herein relate to adapting a language model for automatic speech recognition (ASR) for a new set of words. Instead of retraining the ASR models, language models and grammar models, the system only modifies one grammar model and ensures its compatibility with the existing models in the ASR system.Type: GrantFiled: March 25, 2019Date of Patent: October 12, 2021Assignee: Amazon Technologies, Inc.Inventors: Ankur Gandhe, Ariya Rastrow, Gautam Tiwari, Ashish Vishwanath Shenoy, Chun Chen
-
Publication number: 20210312914Abstract: Described herein is a system for rescoring automatic speech recognition hypotheses for conversational devices that have multi-turn dialogs with a user. The system leverages dialog context by incorporating data related to past user utterances and data related to the system generated response corresponding to the past user utterance. Incorporation of this data improves recognition of a particular user utterance within the dialog.Type: ApplicationFiled: June 7, 2021Publication date: October 7, 2021Inventors: Behnam Hedayatnia, Anirudh Raju, Ankur Gandhe, Chandra Prakash Khatri, Ariya Rastrow, Anushree Venkatesh, Arindam Mandal, Raefer Christopher Gabriel, Ahmad Shikib Mehri
-
Patent number: 11043214Abstract: Described herein is a system for rescoring automatic speech recognition hypotheses for conversational devices that have multi-turn dialogs with a user. The system leverages dialog context by incorporating data related to past user utterances and data related to the system generated response corresponding to the past user utterance. Incorporation of this data improves recognition of a particular user utterance within the dialog.Type: GrantFiled: November 29, 2018Date of Patent: June 22, 2021Assignee: Amazon Technologies, Inc.Inventors: Behnam Hedayatnia, Anirudh Raju, Ankur Gandhe, Chandra Prakash Khatri, Ariya Rastrow, Anushree Venkatesh, Arindam Mandal, Raefer Christopher Gabriel, Ahmad Shikib Mehri
-
Patent number: 10943583Abstract: A system to perform automatic speech recognition (ASR) using a dynamic language model. Portions of the language model can include a group of probabilities rather than a single probability. At runtime individual probabilities of the group are weighted and combined to create an adjusted probability for the portion of the language model. The adjusted probability can be used for ASR processing. The weights can be determined based on a characteristic of the utterance, for example an associated speechlet/application, the specific user speaking, or other characteristic. By applying the weights at runtime the system can use a single language model to dynamically adjust to different utterance conditions.Type: GrantFiled: March 23, 2018Date of Patent: March 9, 2021Assignee: Amazon Technologies, Inc.Inventors: Ankur Gandhe, Ariya Rastrow, Shaswat Pratap Shah
-
Patent number: 10210862Abstract: Neural networks may be used in certain automatic speech recognition systems. To improve performance at these neural networks, the present system converts the lattice into a matrix form, thus maintaining certain information included in the lattice that might otherwise be lost while also placing the lattice in a form that may be manipulated by other components to perform operations such as checking ASR results. The matrix representation of the lattice may be transformed into a vector representation by calculations performed at a recurrent neural network (RNN). By representing the lattice as a vector representation the system may perform additional operations, such as ASR results confirmation.Type: GrantFiled: April 6, 2016Date of Patent: February 19, 2019Assignee: Amazon Technologies, Inc.Inventors: Faisal Ladhak, Ankur Gandhe, Markus Dreyer, Ariya Rastrow, Björn Hoffmeister, Lambert Mathias
-
Patent number: 10176802Abstract: An automatic speech recognition (ASR) system may convert an ASR output lattice into a matrix form, thus maintaining certain information included in the lattice that might otherwise be lost in an N-best list output. The matrix representation of the lattice may be encoded using a recurrent neural network (RNN) to create a vector representation of the lattice. The vector representation may then be used by the system to perform additional operations, such as ASR results confirmation.Type: GrantFiled: April 6, 2016Date of Patent: January 8, 2019Assignee: Amazon Technologies, Inc.Inventors: Faisal Ladhak, Ankur Gandhe, Markus Dreyer, Ariya Rastrow, Björn Hoffmeister, Lambert Mathias
-
Patent number: 10121467Abstract: A language model for automatic speech processing, such as a finite state transducer (FST) may be configured to incorporate information about how a particular word sequence (N-gram) may be used in a similar manner from another N-gram. A score of a component of the FST (such as an arc or state) relating to the first N-gram may be based on information of the second N-gram. Further, the FST may be configured to have an arc between a state of the first N-gram and a state of the second N-gram to allow for cross N-gram back off, rather than backoff from a larger N-gram to a smaller N-gram during traversal of the FST during speech processing.Type: GrantFiled: June 30, 2016Date of Patent: November 6, 2018Assignee: Amazon Technologies, Inc.Inventors: Ankur Gandhe, Denis Sergeyevich Filimonov, Ariya Rastrow, Björn Hoffmeister
-
Patent number: 10032463Abstract: An automatic speech recognition (“ASR”) system produces, for particular users, customized speech recognition results by using data regarding prior interactions of the users with the system. A portion of the ASR system (e.g., a neural-network-based language model) can be trained to produce an encoded representation of a user's interactions with the system based on, e.g., transcriptions of prior utterances made by the user. This user-specific encoded representation of interaction history is then used by the language model to customize ASR processing for the user.Type: GrantFiled: December 29, 2015Date of Patent: July 24, 2018Assignee: Amazon Technologies, Inc.Inventors: Ariya Rastrow, Nikko Ström, Spyridon Matsoukas, Markus Dreyer, Ankur Gandhe, Denis Sergeyevich Filimonov, Julian Chan, Rohit Prasad
-
Patent number: 9569530Abstract: Extracting and mining of quote data across multiple languages, including: retrieving, from a plurality of quote sources, a plurality of commentary summarizations, wherein each commentary summarization is embodied as a machine-readable data structure and wherein the plurality of commentary summarizations include information in at least two or more languages; for each commentary summarization: identifying, within the commentary summarization, quote data, wherein the quote data represents a quote from a commentator; creating a quote tuple for the quote data, the quote tuple including information associated with quantifiable aspects of the quote data; and storing, in a quote tuple repository, the quote tuple; mining, for quote analysis information, the quote tuple repository; and presenting, to a user, the quote analysis information.Type: GrantFiled: November 12, 2013Date of Patent: February 14, 2017Assignee: International Business Machines CorporationInventors: Ankur Gandhe, Amit Kumar R. Singh, Karthik Visweswariah
-
Patent number: 9558269Abstract: Extracting and mining of quote data across multiple languages, including: retrieving, from a plurality of quote sources, a plurality of commentary summarizations, wherein each commentary summarization is embodied as a machine-readable data structure and wherein the plurality of commentary summarizations include information in at least two or more languages; for each commentary summarization: identifying, within the commentary summarization, quote data, wherein the quote data represents a quote from a commentator; creating a quote tuple for the quote data, the quote tuple including information associated with quantifiable aspects of the quote data; and storing, in a quote tuple repository, the quote tuple; mining, for quote analysis information, the quote tuple repository; and presenting, to a user, the quote analysis information.Type: GrantFiled: August 25, 2015Date of Patent: January 31, 2017Assignee: International Business Machines CorporationInventors: Ankur Gandhe, Amit Kumar R. Singh, Karthik Visweswariah
-
Publication number: 20150363487Abstract: Extracting and mining of quote data across multiple languages, including: retrieving, from a plurality of quote sources, a plurality of commentary summarizations, wherein each commentary summarization is embodied as a machine-readable data structure and wherein the plurality of commentary summarizations include information in at least two or more languages; for each commentary summarization: identifying, within the commentary summarization, quote data, wherein the quote data represents a quote from a commentator; creating a quote tuple for the quote data, the quote tuple including information associated with quantifiable aspects of the quote data; and storing, in a quote tuple repository, the quote tuple; mining, for quote analysis information, the quote tuple repository; and presenting, to a user, the quote analysis information.Type: ApplicationFiled: August 25, 2015Publication date: December 17, 2015Inventors: ANKUR GANDHE, AMIT KUMAR R. SINGH, KARTHIK VISWESWARIAH
-
Patent number: 9189967Abstract: Methods and arrangements for enhancing content in discussion forums. Access to an online discussion is provided. A posting by an author participating in the discussion is accepted, and a recommendation is automatically produced for the author for amending the posting to increase the likelihood of response to the posting by other individuals participating in the discussion.Type: GrantFiled: August 31, 2012Date of Patent: November 17, 2015Assignee: International Business Machines CorporationInventors: Amit K. Singh, Rose Catherine Kanjirathinkal, Sachindra Joshi, Ankur Gandhe, Karthik Vesweswariah
-
Patent number: 9189965Abstract: Methods and arrangements for enhancing content in discussion forums. Access to an online discussion is provided. A posting by an author participating in the discussion is accepted, and a recommendation is automatically produced for the author for amending the posting to increase the likelihood of response to the posting by other individuals participating in the discussion.Type: GrantFiled: June 29, 2012Date of Patent: November 17, 2015Assignee: International Business Machines CorporationInventors: Amit K. Singh, Rose Catherine Kanjirathinkal, Sachindra Joshi, Ankur Gandhe, Karthik Visweswariah
-
Publication number: 20150134656Abstract: Extracting and mining of quote data across multiple languages, including: retrieving, from a plurality of quote sources, a plurality of commentary summarizations, wherein each commentary summarization is embodied as a machine-readable data structure and wherein the plurality of commentary summarizations include information in at least two or more languages; for each commentary summarization: identifying, within the commentary summarization, quote data, wherein the quote data represents a quote from a commentator; creating a quote tuple for the quote data, the quote tuple including information associated with quantifiable aspects of the quote data; and storing, in a quote tuple repository, the quote tuple; mining, for quote analysis information, the quote tuple repository; and presenting, to a user, the quote analysis information.Type: ApplicationFiled: November 12, 2013Publication date: May 14, 2015Applicant: International Business Machines CorporationInventors: ANKUR GANDHE, AMIT KUMAR R. SINGH, KARTHIK VISWESWARIAH