Trial-based calibration for audio-based identification, recognition, and detection system
Abstract
The disclosed technologies include methods for generating a calibration model using data that is selected to match the conditions of a particular trial that involves an automated comparison of data samples, such as a comparison-based trial performed by an audio-based recognition, identification, or detection system. The disclosed technologies also include improved methods for selecting candidate data used to build the calibration model. The disclosed technologies further include methods for evaluating the performance of the calibration model and for rejecting a trial when not enough matched candidate data is available to build the calibration model. The disclosed technologies additionally include the use of regularization and automated data generation techniques to further improve the robustness of the calibration model.
- Inventors:
- Issue Date:
- Research Org.:
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States); SRI International, Menlo Park, CA (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 2293890
- Patent Number(s):
- 11823658
- Application Number:
- 16/122,531
- Assignee:
- SRI International (Menlo Park, CA)
- DOE Contract Number:
- AC04-94AL85000
- Resource Type:
- Patent
- Resource Relation:
- Patent File Date: 09/05/2018
- Country of Publication:
- United States
- Language:
- English
Citation Formats
McLaren, Mitchell Leigh, and Lawson, Aaron. Trial-based calibration for audio-based identification, recognition, and detection system. United States: N. p., 2023.
Web.
McLaren, Mitchell Leigh, & Lawson, Aaron. Trial-based calibration for audio-based identification, recognition, and detection system. United States.
McLaren, Mitchell Leigh, and Lawson, Aaron. Tue .
"Trial-based calibration for audio-based identification, recognition, and detection system". United States. https://www.osti.gov/servlets/purl/2293890.
@article{osti_2293890,
title = {Trial-based calibration for audio-based identification, recognition, and detection system},
author = {McLaren, Mitchell Leigh and Lawson, Aaron},
abstractNote = {The disclosed technologies include methods for generating a calibration model using data that is selected to match the conditions of a particular trial that involves an automated comparison of data samples, such as a comparison-based trial performed by an audio-based recognition, identification, or detection system. The disclosed technologies also include improved methods for selecting candidate data used to build the calibration model. The disclosed technologies further include methods for evaluating the performance of the calibration model and for rejecting a trial when not enough matched candidate data is available to build the calibration model. The disclosed technologies additionally include the use of regularization and automated data generation techniques to further improve the robustness of the calibration model.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2023},
month = {11}
}
Works referenced in this record:
Device, System, and Method of Liveness Detection Utilizing Voice Biometrics
patent-application, May 2010
- Aley-Raz, Almog; Krause, Nir Moshe; Salmon, Michael Itzhak
- US Patent Application 12/591636; 20100131273
Joint speaker authentication and key phrase identification
patent, November 2019
- McLaren, Mitchell Leigh; Lawson, Aaron
- US Patent Document 10,476,872
Neural Network-Based Speech Processing
patent-application, April 2013
- Stolcke, Andreas; Slaney, Malcolm; Yella, Sree Harsha
- US Patent Application 14/504722; 20160098987
Fast Speaker Recognition Scoring Using I-Vector Posteriors and Probabilistic Linear Discriminant Analysis
patent-application, February 2016
- Cumani, Sandro; Vair, Claudio; Colibro, Daniele Ernesto
- US Patent Application 14/454169; 20160042739
i‐vector representation based on bottleneck features for language identification
journal, November 2013
- Song, Yan; Jiang, Bing; Bao, YeBo
- Electronics Letters, Vol. 49, Issue 24
Device Access Using Voice Authentication
patent-application, April 2016
- Cheyer, Adam John
- US Patent Application 14/987580; 20160119338
Device Access Using Voice Authentication
patent-application, September 2012
- Cheyer, Adam J.
- US Patent Application 13/053144; 20120245941
Advances in deep neural network approaches to speaker recognition
conference, April 2015
- McLaren, Mitchell; Lei, Yun; Ferrer, Luciana
- 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Multi-Sample Conversational Voice Verification
patent-application, December 2013
- Scheffer, Nicolas; Lei, Yun; Bercow, Douglas A.
- US Patent Application 13/560254; 20130339018
Content-Aware Speaker Recognition
patent-application, April 2015
- Scheffer, Nicolas; Lei, Yun
- US Patent Application 14/264916; 20150112684
System and Method for Dynamic Facial Features for Speaker Recognition
patent-application, March 2015
- Syrdal, Ann K.; Chopra, Sumit; Haffner, PAtrick
- US Patent Application 14/551907; 20150081302
Quality Measure Functions for Calibration of Speaker Recognition Systems in Various Duration Conditions
journal, November 2013
- Mandasari, Miranti Indar; Saeidi, Rahim; McLaren, Mitchell
- IEEE Transactions on Audio, Speech, and Language Processing, Vol. 21, Issue 11
Method and apparatus for acoustic signal characterization
patent, November 2016
- Scheffer, Nicolas; Ferrer, Luciana
- US Patent Document 9,489,965
System and method for providing programming content in response to an audio signal
patent, March 2006
- Murdock, Michael; Pearson, John C.; Sajda, Paul
- US Patent Document 7,013,283
Quality measures based calibration with duration and noise dependency for speaker recognition
journal, September 2015
- Mandasari, Miranti Indar; Saeidi, Rahim; van Leeuwen, David A.
- Speech Communication, Vol. 72
I-vectors in the context of phonetically-constrained short utterances for speaker verification
conference, March 2012
- Larcher, Anthony; Bousquet, Pierre-Michel; Lee, Kong Aik
- 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Method Apparatus and System for Capturing and Analyzing Interacted Based Content
patent-application, December 2004
- Freedman, Ilan; Dolev, Yair; Falik, Talia
- US Patent Application 10/484107; 20040249650