Patents by Inventor Aaron Edwards
Aaron Edwards 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: 7904294Abstract: An automatic speech recognition (ASR) system and method is provided for controlling the recognition of speech utterances generated by an end user operating a communications device. The ASR system and method can be used with a mobile device that is used in a communications network. The ASR system can be used for ASR of speech utterances input into a mobile device, to perform compensating techniques using at least one characteristic and for updating an ASR speech recognizer associated with the ASR system by determined and using a background noise value and a distortion value that is based on the features of the mobile device. The ASR system can be used to augment a limited data input capability of a mobile device, for example, caused by limited input devices physically located on the mobile device.Type: GrantFiled: April 9, 2007Date of Patent: March 8, 2011Assignee: AT&T Intellectual Property II, L.P.Inventors: Richard C. Rose, Sarangarajan Pathasarathy, Aaron Edward Rosenberg, Shrikanth Sambasivan Narayanan
-
Publication number: 20100205124Abstract: Support vector machines are used to classify data contained within a structured dataset such as a plurality of signals generated by a spectral analyzer. The signals are pre-processed to ensure alignment of peaks across the spectra. Similarity measures are constructed to provide a basis for comparison of pairs of samples of the signal. A support vector machine is trained to discriminate between different classes of the samples. to identify the most predictive features within the spectra. In a preferred embodiment feature selection is performed to reduce the number of features that must be considered.Type: ApplicationFiled: February 4, 2010Publication date: August 12, 2010Applicant: HEALTH DISCOVERY CORPORATIONInventors: Asa Ben-Hur, Andre Elisseeff, Olivier Chapelle, Jason Aaron Edward Weston
-
Publication number: 20100166157Abstract: The invention provides a system and method for indexing and organizing voice mail message by the speaker of the message. One or more speaker models are created from voice mail messages received. As additional messages are left, each of the new messages are compared with existing speaker models to determine the identity of the callers of each of the new messages. The voice mail messages are organized within a user's mailbox by caller. Unknown callers may be identified and tagged by the user and then used to create new speaker models and/or update existing speaker models.Type: ApplicationFiled: December 29, 2009Publication date: July 1, 2010Applicant: AT&T Corp.Inventors: Julia Hirschberg, Sarangarajan Parthasarathy, Aaron Edward Rosenberg, Stephen Whittaker
-
Publication number: 20100109831Abstract: A multi-turn coil formed from a single sheet of conductive material and the method of forming same eliminates the use of a weld. The multi-turn coil includes a single sheet of conductive material having at least a first turn in a first plane, and at least a second turn in a second plane, where the first plane is parallel to the second plane. An interconnecting fold interconnects the first and second turns, and any additional turns. The method of forming a multiple turn coil includes providing a continuous strip of conductive material having at least first and second turns extending through substantially 360° and formed in a first plane. The method further includes displacing at least the first turn from the first plane into generally overlapping, parallel relation with the second turn.Type: ApplicationFiled: October 31, 2008Publication date: May 6, 2010Inventors: Andrew Lawrence Podevels, Gary Robert Allen, Aaron Edward Franczyk, Glenn Howard Kuenzler, Derek Lee Watkins, Joshua Ian Rintamaki, Jianwu Li
-
Patent number: 7676442Abstract: Support vector machines are used to classify data contained within a structured dataset such as a plurality of signals generated by a spectral analyzer. The signals are pre-processed to ensure alignment of peaks across the spectra. Similarity measures are constructed to provide a basis for comparison of pairs of samples of the signal. A support vector machine is trained to discriminate between different classes of the samples. to identify the most predictive features within the spectra. In a preferred embodiment feature selection is performed to reduce the number of features that must be considered.Type: GrantFiled: October 30, 2007Date of Patent: March 9, 2010Assignee: Health Discovery CorporationInventors: Asa Ben-Hur, André Elisseeff, Olivier Chapelle, Jason Aaron Edward Weston
-
Patent number: 7664636Abstract: The invention provides a system and method for indexing and organizing voice mail message by the speaker of the message. One or more speaker models are created from voice mail messages received. As additional messages are left, each of the new messages are compared with existing speaker models to determine the identity of the callers of each of the new messages. The voice mail messages are organized within a user's mailbox by caller. Unknown callers may be identified and tagged by the user and then used to create new speaker models and/or update existing speaker models.Type: GrantFiled: April 17, 2000Date of Patent: February 16, 2010Assignee: AT&T Intellectual Property II, L.P.Inventors: Julia Hirschberg, Sarangarajan Parthasarathy, Aaron Edward Rosenberg, Stephen Whittaker
-
Patent number: 7624074Abstract: In a pre-processing step prior to training a learning machine, pre-processing includes reducing the quantity of features to be processed using feature selection methods selected from the group consisting of recursive feature elimination (RFE), minimizing the number of non-zero parameters of the system (l0-norm minimization), evaluation of cost function to identify a subset of features that are compatible with constraints imposed by the learning set, unbalanced correlation score and transductive feature selection. The features remaining after feature selection are then used to train a learning machine for purposes of pattern classification, regression, clustering and/or novelty detection.Type: GrantFiled: October 30, 2007Date of Patent: November 24, 2009Assignee: Health Discovery CorporationInventors: Jason Aaron Edward Weston, Andre′ Elisseeff, Bernard Schoelkopf, Fernando Pérez-Cruz
-
Patent number: 7617163Abstract: Support vector machines are used to classify data contained within a structured dataset such as a plurality of signals generated by a spectral analyzer. The signals are pre-processed to ensure alignment of peaks across the spectra. Similarity measures are constructed to provide a basis for comparison of pairs of samples of the signal. A support vector machine is trained to discriminate between different classes of the samples. to identify the most predictive features within the spectra. In a preferred embodiment feature selection is performed to reduce the number of features that must be considered.Type: GrantFiled: October 9, 2002Date of Patent: November 10, 2009Assignee: Health Discovery CorporationInventors: Asa Ben-Hur, André Elisseeff, Olivier Chapelle, Jason Aaron Edward Weston
-
Patent number: 7542947Abstract: The data mining platform comprises a plurality of system modules, each formed from a plurality of components. Each module has an input data component, a data analysis engine for processing the input data, an output data component for outputting the results of the data analysis, and a web server to access and monitor the other modules within the unit and to provide communication to other units. Each module processes a different type of data, for example, a first module processes microarray (gene expression) data while a second module processes biomedical literature on the Internet for information supporting relationships between genes and diseases and gene functionality. In the preferred embodiment, the data analysis engine is a kernel-based learning machine, and in particular, one or more support vector machines (SVMs).Type: GrantFiled: October 30, 2007Date of Patent: June 2, 2009Assignee: Health Discovery CorporationInventors: Isabelle Guyon, Edward P. Reiss, René Doursat, Jason Aaron Edward Weston, David D. Lewis
-
Publication number: 20090063144Abstract: An automatic speech recognition (ASR) system and method is provided for controlling the recognition of speech utterances generated by an end user operating a communications device. The ASR system and method can be used with a communications device that is used in a communications network. The ASR system can be used for ASR of speech utterances input into a mobile device, to perform compensating techniques using at least one characteristic and for updating an ASR speech recognizer associated with the ASR system by determined and using a background noise value and a distortion value that is based on the features of the mobile device. The ASR system can be used to augment a limited data input capability of a mobile device, for example, caused by limited input devices physically located on the mobile device.Type: ApplicationFiled: November 4, 2008Publication date: March 5, 2009Applicant: AT&T Corp.Inventors: Richard C. Rose, Sarangarajan Pathasarathy, Aaron Edward Rosenberg, Shrikanth Sambasivan Narayanan
-
Publication number: 20090006088Abstract: Speech recognition models are dynamically re-configurable based on user information, application information, background information such as background noise and transducer information such as transducer response characteristics to provide users with alternate input modes to keyboard text entry. Word recognition lattices are generated for each data field of an application and dynamically concatenated into a single word recognition lattice. A language model is applied to the concatenated word recognition lattice to determine the relationships between the word recognition lattices and repeated until the generated word recognition lattices are acceptable or differ from a predetermined value only by a threshold amount. These techniques of dynamic re-configurable speech recognition provide for deployment of speech recognition on small devices such as mobile phones and personal digital assistants as well environments such as office, home or vehicle while maintaining the accuracy of the speech recognition.Type: ApplicationFiled: September 9, 2008Publication date: January 1, 2009Applicant: AT&T Corp.Inventors: Bojana Gajic, Shrikanth Sambasivan Narayanan, Sarangarajan Parthasarathy, Richard Cameron Rose, Aaron Edward Rosenberg
-
Patent number: 7451081Abstract: Speech recognition models are dynamically re-configurable based on user information, application information, background information such as background noise and transducer information such as transducer response characteristics to provide users with alternate input modes to keyboard text entry. Word recognition lattices are generated for each data field of an application and dynamically concatenated into a single word recognition lattice. A language model is applied to the concatenated word recognition lattice to determine the relationships between the word recognition lattices and repeated until the generated word recognition lattices are acceptable or differ from a predetermined value only by a threshold amount. These techniques of dynamic re-configurable speech recognition provide for deployment of speech recognition on small devices such as mobile phones and personal digital assistants as well environments such as office, home or vehicle while maintaining the accuracy of the speech recognition.Type: GrantFiled: March 13, 2007Date of Patent: November 11, 2008Assignee: AT&T Corp.Inventors: Bojana Gajic, Shrikanth Sambasivan Narayanan, Sarangarajan Parthasarathy, Richard Cameron Rose, Aaron Edward Rosenberg
-
Patent number: 7451085Abstract: An automatic speech recognition (ASR) system and method is provided for controlling the recognition of speech utterances generated by an end user operating a communications device. The ASR system and method can be used with a communications device that is used in a communications network. The ASR system can be used for ASR of speech utterances input into a mobile device, to perform compensating techniques using at least one characteristic and for updating an ASR speech recognizer associated with the ASR system by determined and using a background noise value and a distortion value that is based on the features of the mobile device. The ASR system can be used to augment a limited data input capability of a mobile device, for example, caused by limited input devices physically located on the mobile device.Type: GrantFiled: October 1, 2001Date of Patent: November 11, 2008Assignee: AT&T Intellectual Property II, L.P.Inventors: Richard C. Rose, Sarangarajan Pathasarathy, Aaron Edward Rosenberg, Shrikanth Sambasivan Narayanan
-
Patent number: 7444308Abstract: The data mining platform comprises a plurality of system modules (500, 550), each formed from a plurality of components. Each module has an input data component (502, 552), a data analysis engine (504, 554) for processing the input data, an output data component (506, 556) for outputting the results of the data analysis, and a web server (510) to access and monitor the other modules within the unit and to provide communication to other units. Each module processes a different type of data, for example, a first module processes microarray (gene expression) data while a second module processes biomedical literature on the Internet for information supporting relationships between genes and diseases and gene functionality.Type: GrantFiled: June 17, 2002Date of Patent: October 28, 2008Assignee: Health Discovery CorporationInventors: Isabelle Guyon, Edward P. Reiss, René Doursat, Jason Aaron Edward Weston
-
Publication number: 20080215513Abstract: In a pre-processing step prior to training a learning machine, pre-processing includes reducing the quantity of features to be processed using feature selection methods selected from the group consisting of recursive feature elimination (RFE), minimizing the number of non-zero parameters of the system (l0-norm minimization), evaluation of cost function to identify a subset of features that are compatible with constraints imposed by the learning set, unbalanced correlation score and transductive feature selection. The features remaining after feature selection are then used to train a learning machine for purposes of pattern classification, regression, clustering and/or novelty detection.Type: ApplicationFiled: October 30, 2007Publication date: September 4, 2008Inventors: Jason Aaron Edward Weston, Andre' Elisseeff, Bernard Schoelkopf, Fernando Perez-Cruz
-
Patent number: 7406606Abstract: An apparatus, a method, and a computer program are provided for distinguishing relevant security threats. With conventional computer systems, distinguishing security threats from actual security threats is a complex and difficult task because of the general inability to quantify a “threat.” By the use of an intelligent conceptual clustering technique, threats can be accurately distinguished from benign behaviors. Thus, electronic commerce, and Information Technology systems generally, can be made safer without sacrificing efficiency.Type: GrantFiled: April 8, 2004Date of Patent: July 29, 2008Assignee: International Business Machines CorporationInventors: Anil Jagdish Chawla, David Perry Greene, Klaus Julisch, Aaron Edward Fredrick Rankin, Jonathan Michael Seeber, Rhys Ulerich
-
Patent number: 7318051Abstract: In a pre-processing step prior to training a learning machine, pre-processing includes reducing the quantity of features to be processed using feature selection methods selected from the group consisting of recursive feature elimination (RFE), minimizing the number of non-zero parameters of the system (lo-norm minimization), evaluation of cost function to identify a subset of features that are compatible with constraints imposed by the learning set, unbalanced correlation score and transductive feature selection. The features remaining after feature selection are then used to train a learning machine for purposes of pattern classification, regression, clustering and/or novelty detection. (FIG.Type: GrantFiled: May 20, 2002Date of Patent: January 8, 2008Assignee: Health Discovery CorporationInventors: Jason Aaron Edward Weston, André Elisseeff, Bernhard Schoelkopf, Fernando Pérez-Cruz
-
Patent number: 7295970Abstract: Systems and methods for unsupervised segmentation of multi-speaker speech or audio data by speaker. A front-end analysis is applied to input speech data to obtain feature vectors. The speech data is initially segmented and then clustered into groups of segments that correspond to different speakers. The clusters are iteratively modeled and resegmented to obtain stable speaker segmentations. The overlap between segmentation sets is checked to ensure successful speaker segmentation. Overlapping segments are combined and remodeled and resegmented. Optionally, the speech data is processed to produce a segmentation lattice to maximize the overall segmentation likelihood.Type: GrantFiled: January 24, 2003Date of Patent: November 13, 2007Assignee: AT&T CorpInventors: Allen Louis Gorin, Zhu Liu, Sarangarajan Parthasarathy, Aaron Edward Rosenberg
-
Patent number: 7219058Abstract: An automatic speech recognition (ASR) system and method is provided for controlling the recognition of speech utterances generated by an end user operating a communications device. The ASR system and method can be used with a mobile device that is used in a communications network. The ASR system can be used for ASR of speech utterances input into a mobile device, to perform compensating techniques using at least one characteristic and for updating an ASR speech recognizer associated with the ASR system by determined and using a background noise value and a distortion value that is based on the features of the mobile device. The ASR system can be used to augment a limited data input capability of a mobile device, for example, caused by limited input devices physically located on the mobile device.Type: GrantFiled: October 1, 2001Date of Patent: May 15, 2007Assignee: AT&T Corp.Inventors: Richard C. Rose, Sarangarajan Pathasarathy, Aaron Edward Rosenberg, Shrikanth Sambasivan Narayanan
-
Patent number: 7209880Abstract: Speech recognition models are dynamically re-configurable based on user information, application information, background information such as background noise and transducer information such as transducer response characteristics to provide users with alternate input modes to keyboard text entry. Word recognition lattices are generated for each data field of an application and dynamically concatenated into a single word recognition lattice. A language model is applied to the concatenated word recognition lattice to determine the relationships between the word recognition lattices and repeated until the generated word recognition lattices are acceptable or differ from a predetermined value only by a threshold amount. These techniques of dynamic re-configurable speech recognition provide for deployment of speech recognition on small devices such as mobile phones and personal digital assistants as well environments such as office, home or vehicle while maintaining the accuracy of the speech recognition.Type: GrantFiled: March 6, 2002Date of Patent: April 24, 2007Assignee: AT&T Corp.Inventors: Bojana Gajic, Shrikanth Sambasivan Narayanan, Sarangarajan Parthasarathy, Richard Cameron Rose, Aaron Edward Rosenberg