SPEAKER VERIFICATION SYSTEM USING ACOUSTIC DATA AND NON-ACOUSTIC DATA

• US 20070100608A1
• Filed: 12/21/2005
• Published: 05/03/2007
• Est. Priority Date: 11/21/2000
• Status: Active Grant

• Current Assignee: Lawrence Livermore National Security LLC (Government of the United States of America)
• Original Assignee: Regents of the University of California (University of California)
• Inventors: Ng, Lawrence, Burnett, Greg, Holzrichter, John, Gable, Todd
• Granted Patent: US 7,231,350 B2
• Time in Patent Office: Days
• US Class Current: 704/209
• CPC Class Codes: G10L 17/02 Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction

Abstract

A method and system for speech characterization. One embodiment includes a method for speaker verification which includes collecting data from a speaker, wherein the data comprises acoustic data and non-acoustic data. The data is used to generate a template that includes a first set of “template” parameters. The method further includes receiving a real-time identity claim from a claimant, and using acoustic data and non-acoustic data from the identity claim to generate a second set of parameters. The method further includes comparing the first set of parameters to the set of parameters to determine whether the claimant is the speaker. The first set of parameters and the second set of parameters include at least one purely non-acoustic parameter, including a non-acoustic glottal shape parameter derived from averaging multiple glottal cycle waveforms.

Claims

1. A method for speaker verification, comprising:

obtaining at least one set of test sentences by the speaker using a microphone for acoustic data collection and a glottal electromagnetic micropower sensor for electromagnetic data collection, said electromagnetic data collection related to the motions of the speaker'"'"'s tracheal and glottal tissues during speech production;

creating a plurality of speaker verification parameters by extracting parameters from said set of test sentences by the speaker;

collecting a plurality of data from the speaker using a microphone for acoustic data collection and a glottal electromagnetic micropower sensor for electromagnetic data collection, said electromagnetic data collection related to the motions of the speaker'"'"'s tracheal and glottal tissues during speech production;

using said plurality of data to create a plurality of parameters, comprising a glottal shape parameter (GSP) derived from sensing motion of the tracheal and glottal tissues, said parameters created by generating multiple feature vectors; and

comparing said speaker verification parameters with said parameters for speaker verification.

2. The method of claim 1, wherein creating the plurality of parameters comprises extracting a parameter from each of multiple repetitions of each of a set of test sentences by the speaker.

3. The method of claim 1, further comprising averaging the multiple feature vectors to produce a resultant feature vector.

4. The method of claim 1, wherein the GSP is based on average glottal cycle waveforms.

5. The method of claim 4, wherein the non-consecutive two-glottal cycle waveforms are averaged to produce the GSP.