Patents by Inventor Ilya Dan MELAMED
Ilya Dan MELAMED 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: 10971135Abstract: Systems, methods, and computer-readable storage devices for crowd-sourced data labeling. The system requests a respective response from each of a set of entities. The set of entities includes crowd workers. Next, the system incrementally receives a number of responses from the set of entities until one of an accuracy threshold is reached and m responses are received, wherein the accuracy threshold is based on characteristics of the number of responses. Finally, the system generates an output response based on the number of responses.Type: GrantFiled: July 11, 2019Date of Patent: April 6, 2021Assignee: AT&T INTELLECTUAL PROPERTY I, L.P.Inventors: Jason Williams, Tirso Alonso, Barbara B. Hollister, Ilya Dan Melamed
-
Publication number: 20190333497Abstract: Systems, methods, and computer-readable storage devices for crowd-sourced data labeling. The system requests a respective response from each of a set of entities. The set of entities includes crowd workers. Next, the system incrementally receives a number of responses from the set of entities until one of an accuracy threshold is reached and m responses are received, wherein the accuracy threshold is based on characteristics of the number of responses. Finally, the system generates an output response based on the number of responses.Type: ApplicationFiled: July 11, 2019Publication date: October 31, 2019Inventors: Jason Williams, Tirso Alonso, Barbara B. Hollister, Ilya Dan Melamed
-
Patent number: 10360897Abstract: Systems, methods, and computer-readable storage devices for crowd-sourced data labeling. The system requests a respective response from each of a set of entities. The set of entities includes crowd workers. Next, the system incrementally receives a number of responses from the set of entities until one of an accuracy threshold is reached and m responses are received, wherein the accuracy threshold is based on characteristics of the number of responses. Finally, the system generates an output response based on the number of responses.Type: GrantFiled: December 9, 2016Date of Patent: July 23, 2019Assignee: AT&T INTELLECTUAL PROPERTY I, L.P.Inventors: Jason Williams, Tirso Alonso, Barbara B. Hollister, Ilya Dan Melamed
-
Patent number: 10199035Abstract: Systems, methods, and computer-readable storage devices for performing per-channel automatic speech recognition. An example system configured to practice the method combines a first audio signal of a first speaker in a communication session and a second audio signal from a second speaker in the communication session as a first audio channel and a second audio channel. The system can recognize speech in the first audio channel of the recording using a first model specific to the first speaker, and recognize speech in the second audio channel of the recording using a second model specific to the second speaker, wherein the first model is different from the second model. The system can generate recognized speech as an output from the communication session. The system can identify the models based on identifiers of the speakers, such as a telephone number, an IP address, a customer number, or account number.Type: GrantFiled: November 22, 2013Date of Patent: February 5, 2019Assignee: NUANCE COMMUNICATIONS, INC.Inventors: Ilya Dan Melamed, Andrej Ljolje
-
Patent number: 9536517Abstract: Systems, methods, and computer-readable storage devices for crowd-sourced data labeling. The system requests a respective response from each of a set of entities. The set of entities includes crowd workers. Next, the system incrementally receives a number of responses from the set of entities until one of an accuracy threshold is reached and m responses are received, wherein the accuracy threshold is based on characteristics of the number of responses. Finally, the system generates an output response based on the number of responses.Type: GrantFiled: November 18, 2011Date of Patent: January 3, 2017Assignee: AT&T Intellectual Property I, L.P.Inventors: Jason Williams, Tirso Alonso, Barbara B. Hollister, Ilya Dan Melamed
-
Publication number: 20160078360Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for performing trend analysis of speech. A system practicing the method receives a speech trend analysis request having candidate feature constraints, an objective function with respect to a speech trend to be analyzed, and a set of speech record constraints. The system selects a subset of speech records from the group of speech records based on the set of speech record constraints to yield selected speech records, identifies features in the selected speech records based on the set of candidate feature constraints to yield identified features, and assigns a weight to each of the identified features based on the objective function. Then the system ranks the identified features by their respective weights to yield ranked identified features, and outputs at least one of the ranked identified features associated with a speech-based trend in response to the speech trend analysis request.Type: ApplicationFiled: November 30, 2015Publication date: March 17, 2016Inventors: ILYA Dan MELAMED, Mazin GILBERT
-
Patent number: 9213978Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for performing trend analysis of speech. A system practicing the method receives a speech trend analysis request having candidate feature constraints, an objective function with respect to a speech trend to be analyzed, and a set of speech record constraints. The system selects a subset of speech records from the group of speech records based on the set of speech record constraints to yield selected speech records, identifies features in the selected speech records based on the set of candidate feature constraints to yield identified features, and assigns a weight to each of the identified features based on the objective function. Then the system ranks the identified features by their respective weights to yield ranked identified features, and outputs at least one of the ranked identified features associated with a speech-based trend in response to the speech trend analysis request.Type: GrantFiled: September 30, 2010Date of Patent: December 15, 2015Assignee: AT&T Intellectual Property I, L.P.Inventors: Ilya Dan Melamed, Mazin Gilbert
-
Publication number: 20150149162Abstract: Disclosed herein are systems, methods, and computer-readable storage devices for performing per-channel automatic speech recognition. An example system configured to practice the method combines a first audio signal of a first speaker in a communication session and a second audio signal from a second speaker in the communication session as a first audio channel and a second audio channel. The system can recognize speech in the first audio channel of the recording using a first model associated with the first speaker, and recognize speech in the second audio channel of the recording using a second model associated with the second speaker, wherein the first model is different from the second model. The system can generate recognized speech as an output from the communication session. The system can identify the models based on identifiers of the speakers, such as a telephone number, an IP address, a customer number, or account number.Type: ApplicationFiled: November 22, 2013Publication date: May 28, 2015Applicant: AT&T Intellectual Property I, L.P.Inventors: Ilya Dan MELAMED, Andrej LJOLJE
-
Patent number: 8606585Abstract: A method, apparatus, and computer-readable medium for editing a data stream based on a corpus are provided. The data stream includes stream words. A sequence includes a predetermined number of sequential words of the stream words. The method, apparatus, and computer-readable medium determine whether the sequence exists in the corpus at least at a predetermined minimum frequency. When the sequence exists in the corpus at least at the predetermined minimum frequency, the sequence is edited in the data stream.Type: GrantFiled: September 17, 2010Date of Patent: December 10, 2013Assignee: AT&T Intellectual Property I, L.P.Inventors: Ilya Dan Melamed, Yeon-Jun Kim
-
Publication number: 20130132080Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for crowd-sourced data labeling. The system requests a respective response from each of a set of entities. The set of entities includes crowd workers. Next, the system incrementally receives a number of responses from the set of entities until at least one of an accuracy threshold is reached and m responses are received, wherein the accuracy threshold is based on characteristics of the number of responses. Finally, the system generates an output response based on the number of responses.Type: ApplicationFiled: November 18, 2011Publication date: May 23, 2013Applicant: AT&T Intellectual Property I, L.P.Inventors: Jason Williams, Tirso Alonso, Barbara B. Hollister, Ilya Dan Melamed
-
Publication number: 20120084081Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for performing trend analysis of speech. A system practicing the method receives a speech trend analysis request having candidate feature constraints, an objective function with respect to a speech trend to be analyzed, and a set of speech record constraints. The system selects a subset of speech records from the group of speech records based on the set of speech record constraints to yield selected speech records, identifies features in the selected speech records based on the set of candidate feature constraints to yield identified features, and assigns a weight to each of the identified features based on the objective function. Then the system ranks the identified features by their respective weights to yield ranked identified features, and outputs at least one of the ranked identified features associated with a speech-based trend in response to the speech trend analysis request.Type: ApplicationFiled: September 30, 2010Publication date: April 5, 2012Applicant: AT&T Intellectual Property I, L.P.Inventors: ILYA Dan MELAMED, Mazin Gilbert
-
Publication number: 20110145002Abstract: A method, apparatus, and computer-readable medium for editing a data stream based on a corpus are provided. The data stream includes stream words. A sequence includes a predetermined number of sequential words of the stream words. The method, apparatus, and computer-readable medium determine whether the sequence exists in the corpus at least at a predetermined minimum frequency. When the sequence exists in the corpus at least at the predetermined minimum frequency, the sequence is edited in the data stream.Type: ApplicationFiled: September 17, 2010Publication date: June 16, 2011Applicant: AT&T INTELLECTUAL PROPERTY I, L.P.Inventors: Ilya Dan MELAMED, Yeon-Jun KIM