SPEAKER IDENTIFICATION METHOD AND SPEAKER IDENTIFICATION DEVICE

• US 20160314790A1
• Filed: 04/16/2016
• Published: 10/27/2016
• Est. Priority Date: 04/22/2015
• Status: Active Grant

• Current Assignee: Panasonic Corporation (Panasonic Holdings Corporation)
• Original Assignee: Panasonic Corporation (Panasonic Holdings Corporation)
• Inventors: MATSUI, TOMOKO, TSUJIKAWA, MISAKI
• Granted Patent: US 9,947,324 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G10L 17/04 Training, enrolment or model building
• CPC Class Codes: G10L 17/06 Decision making techniques; Pattern matching strategies

Abstract

A first similarity degree calculation unit calculates first similarity degrees between a feature value in voice signal of each of first speakers and each feature value in a plurality of unspecified speaker models of a plurality of unspecified speakers, a model specifying unit specifies a plurality of the unspecified speaker models for which the first similarity degrees are equal to or greater than a prescribed value, a second speaker model storage unit associates and stores each of the first speakers and the specified unspecified speaker models, a second similarity degree calculation unit calculates, for each of the first speakers, a plurality of second similarity degrees between a feature value in a voice signal of a second speaker and each feature values in the unspecified speaker models associated with the first speakers and stored in the second speaker model storage unit, and a speaker identification unit that specifies the second speaker based on the second similarity degrees.

Claims

1. A speaker identification method that includes:

learning mode processing in which a first database, in which a plurality of unspecified speakers and a plurality of unspecified speaker models obtained by modeling features of voices of the plurality of unspecified speakers are associated and stored, is used to create a second database, in which first speakers who are not stored in the first database and a plurality of the unspecified speaker models are associated and stored; and

identification mode processing in which the second database is used to identify a second speaker,

wherein, in the learning mode processing,

voice signal of each of the first speakers is acquired,

first similarity degrees between a feature value in the acquired voice signal of each of the first speakers and each feature value in the plurality of unspecified speaker models stored in the first database are calculated,

a plurality of the unspecified speaker models for which the calculated first similarity degrees are equal to or greater than a prescribed value are specified,

and each of the first speakers and the specified plurality of unspecified speaker models are associated and stored in the second database,

and in the identification mode processing,

a voice signal of the second speaker is acquired,

a plurality of second similarity degrees between a feature value in the acquired voice signal of the second speaker and each feature value in the plurality of unspecified speaker models associated with the first speakers and stored in the second database are calculated for each of the first speakers,

and one of the first speakers stored in the second database who corresponds to the second speaker is specified based on the calculated plurality of second similarity degrees.

2. The speaker identification method according to claim 1,

wherein a total value obtained by totaling the calculated plurality of second similarity degrees is calculated for each of the first speakers, and the one of the first speakers stored in the second database who corresponds to the second speaker is specified based on the calculated total values.

3. The speaker identification method according to claim 1,

wherein the calculated plurality of second similarity degrees are corrected by multiplying each of the plurality of second similarity degrees by a weighting value that corresponds to a ranking of the first similarity degrees,

a total value obtained by totaling the corrected plurality of second similarity degrees is calculated for each of the first speakers, and the one of the first speakers stored in the second database who corresponds to the second speaker is specified based on the calculated total values.

4. The speaker identification method according to claim 3,

wherein the weighting value increases as the first similarity degrees increase.

5. The speaker identification method according to claim 1,

wherein the calculated plurality of second similarity degrees are corrected by multiplying each of the plurality of second similarity degrees by a first similarity degree,

a total value obtained by totaling the corrected plurality of second similarity degrees is calculated for each of the first speakers, and the one of the first speakers stored in the second database who corresponds to the second speaker is specified based on the calculated total values.

6. The speaker identification method according to claim 1,

wherein a total value obtained by totaling the plurality of second similarity degrees that are equal to or greater than a prescribed value from among the calculated plurality of second similarity degrees is calculated for each of the first speakers, and the one of the first speakers stored in the second database who corresponds to the second speaker is specified based on the calculated total values.

7. The speaker identification method according to claim 2,

wherein the second speaker is specified as the one of the first speakers stored in the second database having the highest calculated total value.

8. The speaker identification method according to claim 1,

wherein, in the learning mode processing,

speaker models corresponding to the first speakers are newly created based on the specified plurality of unspecified speaker models and the acquired voice signals of the first speakers,

and the created speaker models are associated with the first speakers and stored in a third database,

and in the identification mode processing,

for each first speaker, a third similarity degree between a feature value in the acquired voice signal of the second speaker and a feature value in the speaker model associated with the first speaker and stored in the third database is calculated,

and one of the first speakers stored in the third database who corresponds to the second speaker is specified based on the calculated third similarity degrees.

9. The speaker identification method according to claim 8,

wherein, in a case where the second speaker is not specified as being any of the first speakers stored in the third database, the plurality of second similarity degrees between the feature value in the acquired voice signal of the second speaker and each feature value in the plurality of unspecified speaker models associated with the first speakers and stored in the second database are calculated for each of the first speakers,

and the one of the first speakers stored in the second database who corresponds to the second speaker is specified based on the calculated plurality of second similarity degrees.

10. The speaker identification method according to claim 1,

wherein, after the identification mode processing has been performed, the first similarity degrees corresponding to each of the unspecified speaker models calculated in the learning mode processing and the second similarity degrees corresponding to each of the unspecified speaker models calculated in the identification mode processing are compared,

and in a case where there is a prescribed number or more of the unspecified speaker models for which a difference between the first similarity degrees and the second similarity degrees is equal to or greater than a prescribed value, the learning mode processing is performed again.

11. The speaker identification method according to claim 1,

wherein, after the identification mode processing has been performed, the first similarity degrees corresponding to each of the unspecified speaker models calculated in the learning mode processing and the second similarity degrees corresponding to each of the unspecified speaker models calculated in the identification mode processing are compared,

and in a case where there is a prescribed number or more of the unspecified speaker models for which a difference between the first similarity degrees and the second similarity degrees is equal to or greater than a prescribed value, the first similarity degrees corresponding to the unspecified speaker models stored in the second database for which the difference is equal to or greater than the prescribed value are amended to the second similarity degrees calculated in the identification mode processing.

12. The speaker identification method according to claim 1,

wherein at least one of the learning mode processing and the identification mode processing is executed by a processor.

13. A speaker identification device comprising:

a learning mode processing unit that uses a first database, in which a plurality of unspecified speakers and a plurality of unspecified speaker models obtained by modeling features of voices of the plurality of unspecified speakers are associated and stored, to create a second database, in which first speakers who are not stored in the first database and a plurality of the unspecified speaker models are associated and stored; and

an identification mode processing unit that uses the second database to identify a second speaker,

wherein, the learning mode processing unit includes

a first voice acquirer that acquires voice signal of each of the first speakers,

a first similarity degree calculator that calculates first similarity degrees between a feature value in the voice signal of each of the first speakers acquired by the first voice acquirer and each feature value in the plurality of unspecified speaker models stored in the first database,

a first specifier that specifies a plurality of the unspecified speaker models for which the first similarity degrees calculated by the first similarity degree calculator are equal to or greater than a prescribed value,

and a storage processing unit that associates and stores each of the first speakers and the plurality of unspecified speaker models specified by the first specifier in the second database,

and the identification mode processing unit includes

a second voice acquirer that acquires a voice signal of the second speaker,

a second similarity degree calculator that calculates, for each of the first speakers, a plurality of second similarity degrees between a feature value in the voice signal of the second speaker acquired by the second voice acquirer and each feature value in the plurality of unspecified speaker models associated with the first speakers and stored in the second database,

and a second specifier that, based on the plurality of second similarity degrees calculated by the second similarity degree calculator, specifies one of the first speakers stored in the second database who corresponds to the second speaker.

14. The speaker identification device according to claim 13,

wherein at least one of the learning mode processing unit and the identification mode processing unit includes a processor.

15. A speaker identification method that includes:

learning mode processing in which a first database, in which a plurality of unspecified speakers and a plurality of unspecified speaker models obtained by modeling features of voices of the plurality of unspecified speakers are associated and stored, is used to create a second database, in which first speakers who are not stored in the first database and a plurality of the unspecified speaker models are associated and stored; and

identification mode processing in which the second database is used to identify a second speaker,

wherein, in the learning mode processing,

voice signal of each of the first speakers is acquired,

first similarity degrees between a feature value in the acquired voice signal of each of the first speakers and each feature value in the plurality of unspecified speaker models of a plurality of the unspecified speakers who are different from the first speakers and are stored in the first database are calculated,

a plurality of the unspecified speaker models for which the calculated first similarity degrees are equal to or greater than a prescribed value are specified,

speaker model corresponding to each of the first speakers is newly created based on the specified plurality of the unspecified speaker models and the acquired voice signals of the first speakers,

and the created speaker model is associated with the first speakers and stored in the second database,

and in the identification mode processing,

a voice signal of the second speaker is acquired,

a plurality of second similarity degrees between a feature value in the acquired voice signal of the second speaker and feature values in the speaker models associated with the first speakers and stored in the second database are calculated for each of the first speakers,

and one of the first speakers stored in the second database who corresponds to the second speaker is specified based on the calculated plurality of second similarity degrees.

16. The speaker identification method according to claim 15,

wherein at least one of the learning mode processing and the identification mode processing is executed by a processor.