Speaker verification system using acoustic data and non-acoustic data
First Claim
1. A method for speaker verification, comprising:
- collecting a plurality of data from a speaker, wherein the plurality of data comprises acoustic data and non-acoustic data;
using the plurality of data to generate a template comprising a first plurality of parameters;
receiving a real-time identity claim from a claimant;
using a plurality of acoustic data and non-acoustic data from the identity claim to generate a second plurality of parameters; and
comparing the first plurality of parameters to the second plurality of parameters to determine whether the claimant is the speaker, wherein the first plurality of parameters and the second plurality of parameters include at least one purely non-acoustic parameter, including a non-acoustic glottal shape parameter parameter derived from averaging multiple glottal cycle waveforms.
5 Assignments
0 Petitions
Accused Products
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.
133 Citations
45 Claims
-
1. A method for speaker verification, comprising:
collecting a plurality of data from a speaker, wherein the plurality of data comprises acoustic data and non-acoustic data; using the plurality of data to generate a template comprising a first plurality of parameters; receiving a real-time identity claim from a claimant; using a plurality of acoustic data and non-acoustic data from the identity claim to generate a second plurality of parameters; and comparing the first plurality of parameters to the second plurality of parameters to determine whether the claimant is the speaker, wherein the first plurality of parameters and the second plurality of parameters include at least one purely non-acoustic parameter, including a non-acoustic glottal shape parameter parameter derived from averaging multiple glottal cycle waveforms. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
-
16. A system for speaker verification, comprising:
-
at least one microphone for collecting acoustic data from a speaker'"'"'s voice; at least one sensor for collecting non-acoustic data from movements of the speaker'"'"'s body; at least one processor; a memory device coupled to the processor, wherein the memory device stores instructions that when executed cause the processor to generate a template using the acoustic data and non-acoustic data, wherein the template comprises a first plurality of parameters, wherein when a claimant speaks an identity claim into the at least one microphone, the instruction further cause the processor to generate a second plurality of parameters, and to compare the first plurality of parameters to the second plurality of parameters to determine whether the claimant is the speaker, wherein the first plurality of parameters and the second plurality of parameters include at least one purely non-acoustic parameter, including a non-acoustic glottal shape parameter parameter derived from averaging multiple glottal cycle waveforms. - View Dependent Claims (17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30)
-
-
31. An electromagnetic medium, having stored thereon instructions that when executed, cause a processor to:
-
collect a plurality of data from a speaker, wherein the plurality of data comprises acoustic data and non-acoustic data; use the plurality of data to generate a template comprising a first plurality of parameters; receive a real-time identity claim from a claimant; use a plurality of acoustic data and non-acoustic data from the identity claim to generate a second plurality of parameters; and compare the first plurality of parameters to the second plurality of parameters to determine whether the claimant is the speaker, wherein the first plurality of parameters and the second plurality of parameters include at least one purely non-acoustic parameter, including a non-acoustic glottal shape parameter derived from averaging multiple glottal cycle waveforms. - View Dependent Claims (32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45)
-
Specification