Patents by Inventor Marsal Gavalda

Marsal Gavalda 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: 9774747
    Abstract: A transcription system automates the control of the playback of the audio to accommodate the user's ability to transcribe the words spoken. In some examples, a delay between playback and typed input is estimated by processing the typed words using a wordspotting approach. The estimated delay is used as in input to an automated speed control, for example, to maintain a target or maximum delay between playback and typed input.
    Type: Grant
    Filed: April 29, 2011
    Date of Patent: September 26, 2017
    Assignee: NEXIDIA INC.
    Inventors: Jacob B. Garland, Marsal Gavalda
  • Patent number: 9697278
    Abstract: In a general aspect, an approach to query specification includes processing a query by identifying one or more terms and validating the terms using a first corpus of media elements. The result of the validation is used to form a modified query. In some examples, identifying the one or more terms is based on a second corpus of media elements, which may include a different type of media than the first corpus. In some examples, the validating of the terms includes accepting input from a user according to parts of the elements of the first corpus presented to the user.
    Type: Grant
    Filed: November 1, 2011
    Date of Patent: July 4, 2017
    Assignee: NEXIDIA INC.
    Inventors: Robert W. Morris, Neeraj Singh Verma, John Willcutts, Marsal Gavalda
  • Patent number: 9536567
    Abstract: In an aspect, in general, method for aligning an audio recording and a transcript includes receiving a transcript including a plurality of terms, each term of the plurality of terms associated with a time location within a different version of the audio recording, forming a plurality of search terms from the terms of the transcript, determining possible time locations of the search terms in the audio recording, determining a correspondence between time locations within the different version of the audio recording associated with the search terms and the possible time locations of the search terms in the audio recording, and aligning the audio recording and the transcript including updating the time location associated with terms of the transcript based on the determined correspondence.
    Type: Grant
    Filed: September 4, 2012
    Date of Patent: January 3, 2017
    Assignee: NEXIDIA INC.
    Inventors: Jacob B. Garland, Drew Lanham, Daryl Kip Watters, Marsal Gavalda, Mark Finlay, Kenneth K. Griggs
  • Patent number: 9361879
    Abstract: In one aspect, a method for processing media includes accepting a query. One or more language patterns are identified that are similar to the query. A putative instance of the query is located in the media. The putative instance is associated with a corresponding location in the media. The media in a vicinity of the putative instance is compared to the identified language patterns and data characterizing the putative instance of the query is provided according to the comparing of the media to the language patterns, for example, as a score for the putative instance that is determined according to the comparing of the media to the language patterns.
    Type: Grant
    Filed: February 24, 2009
    Date of Patent: June 7, 2016
    Assignee: NEXIDIA INC.
    Inventors: Robert W. Morris, Jon A. Arrowood, Mark A. Clements, Kenneth King Griggs, Peter S. Cardillo, Marsal Gavalda
  • Patent number: 9001976
    Abstract: A method for speaker adaptation includes receiving a plurality of media files, each associated with a call center agent of a plurality of call center agents and receiving a plurality of terms. Speech processing is performed on at least some of the media files to identify putative instances of at least some of the plurality of terms. Each putative instance is associated with a hit quality that characterizes a quality of recognition of the corresponding term. One or more call center agents for performing speaker adaptation are determined, including identifying call center agents that are associated with at least one media file that includes one or more putative instances with a hit quality below a predetermined threshold. Speaker adaptation is performed for each identified call center agent based on the media files associated with the identified call center agent and the identified instances of the plurality of terms.
    Type: Grant
    Filed: May 3, 2012
    Date of Patent: April 7, 2015
    Assignee: Nexidia, Inc.
    Inventors: Jon A. Arrowood, Robert W. Morris, Marsal Gavalda
  • Patent number: 8719022
    Abstract: An audio processing system makes use of a number of levels of compression or data reduction, thereby providing reduced storage requirements while maintaining a high accuracy of keyword detection in the original audio input.
    Type: Grant
    Filed: April 29, 2011
    Date of Patent: May 6, 2014
    Assignee: Nexidia Inc.
    Inventors: Jon A. Arrowood, Robert W. Morris, Peter S. Cardillo, Marsal Gavalda
  • Patent number: 8670983
    Abstract: A method for determining a similarity between a first audio source and a second audio source includes: for the first audio source, determining a first frequency of occurrence for each of a plurality of phoneme sequences and determining a first weighted frequency for each of the plurality of phoneme sequences based on the first frequency of occurrence for the phoneme sequence; for the second audio source, determining a second frequency of occurrence for each of a plurality of phoneme sequences and determining a second weighted frequency for each of the plurality of phoneme sequences based on the second frequency of occurrence for the phoneme sequence; comparing the first weighted frequency for each phoneme sequence with the second weighted frequency for the corresponding phoneme sequence; and generating a similarity score representative of a similarity between the first audio source and the second audio source based on the results of the comparing.
    Type: Grant
    Filed: August 30, 2011
    Date of Patent: March 11, 2014
    Assignee: Nexidia Inc.
    Inventors: Jacob B. Garland, Jon A. Arrowood, Drew Lanham, Marsal Gavalda
  • Publication number: 20130294587
    Abstract: A method for speaker adaptation includes receiving a plurality of media files, each associated with a call center agent of a plurality of call center agents and receiving a plurality of terms. Speech processing is performed on at least some of the media files to identify putative instances of at least some of the plurality of terms. Each putative instance is associated with a hit quality that characterizes a quality of recognition of the corresponding term. One or more call center agents for performing speaker adaptation are determined, including identifying call center agents that are associated with at least one media file that includes one or more putative instances with a hit quality below a predetermined threshold. Speaker adaptation is performed for each identified call center agent based on the media files associated with the identified call center agent and the identified instances of the plurality of terms.
    Type: Application
    Filed: May 3, 2012
    Publication date: November 7, 2013
    Applicant: Nexidia Inc.
    Inventors: Jon A. Arrowood, Robert W. Morris, Marsal Gavalda
  • Patent number: 8494133
    Abstract: A method includes accepting, via an input interface, a caller identifier parameter and a target value of at least one call series parameter; identifying, using a data processor, a plurality of calls each associated with the caller identifier parameter from amongst a set of calls stored in a call center database; analyzing, using the data processor, the identified plurality of calls to determine a value of the at least one call series parameter for the identified plurality of calls; comparing, using the data processor, the determined value of the at least one call series parameter with the target value of the at least one call series parameter; and defining, using the data processor, the identified plurality of calls as a call series based at least in part on results of the comparing.
    Type: Grant
    Filed: June 24, 2010
    Date of Patent: July 23, 2013
    Assignee: Nexidia Inc.
    Inventors: Christopher J. Jeffs, Marsal Gavalda, Philip Kyle Pledger, Robert Troy Surdick
  • Publication number: 20130110849
    Abstract: In a general aspect, an approach to query specification includes processing a query by identifying one or more terms and validating the terms using a first corpus of media elements. The result of the validation is used to form a modified query. In some examples, identifying the one or more terms is based on a second corpus of media elements, which may include a different type of media than the first corpus. In some examples, the validating of the terms includes accepting input from a user according to parts of the elements of the first corpus presented to the user.
    Type: Application
    Filed: November 1, 2011
    Publication date: May 2, 2013
    Applicant: Nexidia Inc.
    Inventors: Robert W. Morris, Neeraj Singh Verma, John Willcutts, Marsal Gavalda
  • Patent number: 8429171
    Abstract: An information retrieval system for search a corpus is configured to, for each of a plurality of searches of at least a subset of the corpus, generate a confidence score for each of a plurality of putative occurrences of a search query in the at least a subset of the corpus. The system is further configured to adjust a threshold to maintain a consistent user experience across the plurality of searches according to a consistency criterion. The system is further configured to display putative occurrences of the search query having a confidence score greater than the threshold.
    Type: Grant
    Filed: August 19, 2008
    Date of Patent: April 23, 2013
    Assignee: Nexidia Inc.
    Inventors: Robert W. Morris, Marsal Gavalda
  • Patent number: 8411841
    Abstract: Some general aspects of the invention relate to systems and methods for improving contact center agent performance, for instance, by integrating real-time call monitoring with speech analytics to present agents with information useful to the handling of the current calls. In some implementations, phonetically based speech analysis techniques are applied to process live audio streams to identify key words and/or phrases of relevance, based on which knowledge articles can be selectively presented to agents to drive more efficient business processes.
    Type: Grant
    Filed: July 16, 2010
    Date of Patent: April 2, 2013
    Assignee: Nexidia Inc.
    Inventors: Gordon Edwards, John Willcutts, Jon W. Ezrine, Marsal Gavalda
  • Publication number: 20130060572
    Abstract: In an aspect, in general, method for aligning an audio recording and a transcript includes receiving a transcript including a plurality of terms, each term of the plurality of terms associated with a time location within a different version of the audio recording, forming a plurality of search terms from the terms of the transcript, determining possible time locations of the search terms in the audio recording, determining a correspondence between time locations within the different version of the audio recording associated with the search terms and the possible time locations of the search terms in the audio recording, and aligning the audio recording and the transcript including updating the time location associated with terms of the transcript based on the determined correspondence.
    Type: Application
    Filed: September 4, 2012
    Publication date: March 7, 2013
    Applicant: Nexidia Inc.
    Inventors: Jacob B. Garland, Drew Lanham, Daryl Kip Watters, Marsal Gavalda, Mark Finlay, Kenneth K. Griggs
  • Publication number: 20130035936
    Abstract: A transcription system is applicable to transcription for a language in which there is limited pronunciation and/or acoustic data. A transcription station is configured using pronunciation data and acoustic data for use with the language. The pronunciation data and/or the acoustic data is initially from another dialect of a language, another language from a language group, or is universal (e.g., not specific to any particular language). A partial transcription of the audio recording is accepted via the transcription station (e.g., from a transcriptionist). One or more repetitions of one or more portions of the partial transcription are identified in the audio recording, and can be accepted during transcription. The pronunciation data and/or the acoustic data is updated in a bootstrapping manner during transcription, thereby improving the efficiency of the transcription process.
    Type: Application
    Filed: August 1, 2012
    Publication date: February 7, 2013
    Applicant: Nexidia Inc.
    Inventors: Jacob B. Garland, Marsal Gavalda
  • Publication number: 20120284026
    Abstract: In an aspect, in general, a method for computer assisted speaker authentication in a voice communication session includes establishing a voice communication session between a first speaker and an agent, accepting a first voice signal from the first speaker, determining a voice characteristic measure of the first voice signal, including characterizing a similarity of the first voice signal to each of one or more stored characterizations of voice signals previously acquired from one or more known speakers, and providing an interface to the agent during the voice communication session between the agent and the first speaker, including presenting an indicator based on the determined voice characteristic measure to the agent.
    Type: Application
    Filed: May 6, 2011
    Publication date: November 8, 2012
    Applicant: Nexidia Inc.
    Inventors: Peter S. Cardillo, Marsal Gavalda
  • Publication number: 20120278071
    Abstract: A transcription system automates the control of the playback of the audio to accommodate the user's ability to transcribe the words spoken. In some examples, a delay between playback and typed input is estimated by processing the typed words using a wordspotting approach. The estimated delay is used as in input to an automated speed control, for example, to maintain a target or maximum delay between playback and typed input.
    Type: Application
    Filed: April 29, 2011
    Publication date: November 1, 2012
    Applicant: Nexidia Inc.
    Inventors: Jacob B. Garland, Marsal Gavalda
  • Publication number: 20120278079
    Abstract: An audio processing system makes use of a number of levels of compression or data reduction, thereby providing reduced storage requirements while maintaining a high accuracy of keyword detection in the original audio input.
    Type: Application
    Filed: April 29, 2011
    Publication date: November 1, 2012
    Inventors: Jon A. Arrowood, Robert W. Morris, Peter S. Cardillo, Marsal Gavalda
  • Publication number: 20120059656
    Abstract: A method for determining a similarity between a first audio source and a second audio source includes: for the first audio source, determining a first frequency of occurrence for each of a plurality of phoneme sequences and determining a first weighted frequency for each of the plurality of phoneme sequences based on the first frequency of occurrence for the phoneme sequence; for the second audio source, determining a second frequency of occurrence for each of a plurality of phoneme sequences and determining a second weighted frequency for each of the plurality of phoneme sequences based on the second frequency of occurrence for the phoneme sequence; comparing the first weighted frequency for each phoneme sequence with the second weighted frequency for the corresponding phoneme sequence; and generating a similarity score representative of a similarity between the first audio source and the second audio source based on the results of the comparing.
    Type: Application
    Filed: August 30, 2011
    Publication date: March 8, 2012
    Applicant: Nexidia Inc.
    Inventors: Jacob B. Garland, Jon A. Arrowood, Drew Lanham, Marsal Gavalda
  • Publication number: 20120010736
    Abstract: A method for detecting sections of a known input in an unknown input includes processing the known input to form a series of discrete-valued feature values associated with corresponding time locations in the known input. Index data associating a plurality of the feature values each with one or more time locations in the known input is then formed. The unknown input is processed to form a series of discrete-valued features values. A time offset between the unknown input and the known input is determined by determining time locations in the known input associated with the feature values of the unknown input. Determining the time offset may include maintaining a distribution of time offsets based on successive determined time locations of the feature values of the unknown input.
    Type: Application
    Filed: July 9, 2010
    Publication date: January 12, 2012
    Applicant: Nexidia Inc.
    Inventors: Peter S. Cardillo, Marsal Gavalda
  • Patent number: 8051086
    Abstract: Some general aspects of the invention relate to systems and methods of processing data, for example, to improve customer interactions. One aspect, in particular, relates to a computer-implemented method that includes accepting user input for analysis of a database having media data and metadata. The media data includes a group of audio recordings and the metadata includes descriptive information of the group of audio recordings. A representation of a set of call series is formed based on user input, and processed to generate an analysis report. A visual representation of the analysis report is formed for presentation to a user.
    Type: Grant
    Filed: June 24, 2009
    Date of Patent: November 1, 2011
    Assignee: Nexidia Inc.
    Inventors: Christopher J. Jeffs, Marsal Gavalda, Philip Kyle Pledger, Robert Troy Surdick