Dynamic threshold for speaker verification

• US 9,679,569 B2
• Filed: 11/03/2016
• Issued: 06/13/2017
• Est. Priority Date: 06/24/2014
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Foerster, Jakob Nicolaus, Melendo Casado, Diego
• Primary Examiner(s): Baker, Charlotte M
• Application Number:
  US15/342,388
• Publication Number:
  US 20170103759A1
• Time in Patent Office: 222 Days
• Field of Search: 704250
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques
• 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
• CPC Class Codes: G10L 17/04 Training, enrolment or model building
• CPC Class Codes: G10L 17/06 Decision making techniques; Pattern matching strategies
• CPC Class Codes: G10L 17/08 Use of distortion metrics or a particular distance between probe pattern and reference templates
• CPC Class Codes: G10L 17/12 Score normalisation
• CPC Class Codes: G10L 17/14 Use of phonemic categorisation or speech recognition prior to speaker recognition or verification
• CPC Class Codes: G10L 17/20 Pattern transformations or operations aimed at increasing system robustness, e.g. against channel noise or different working conditions
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 17/24 the user being prompted to utter a password or a predefined phrase
• CPC Class Codes: G10L 25/84 for discriminating voice from noise
• CPC Class Codes: H04M 3/385 using speech signals

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for a dynamic threshold for speaker verification are disclosed. In one aspect, a method includes the actions of receiving, for each of multiple utterances of a hotword, a data set including at least a speaker verification confidence score, and environmental context data. The actions further include selecting from among the data sets, a subset of the data sets that are associated with a particular environmental context. The actions further include selecting a particular data set from among the subset of data sets based on one or more selection criteria. The actions further include selecting, as a speaker verification threshold for the particular environmental context, the speaker verification confidence score. The actions further include providing the speaker verification threshold for use in performing speaker verification of utterances that are associated with the particular environmental context.

Claims

1. A computer-implemented method comprising:

receiving, by a computing device that uses voice-based speaker identification, data identifying a user;

prompting the user to confirm whether a previously received utterance was properly identified, properly rejected, falsely identified, or falsely rejected using voice-based speaker identification;

receiving, from the user, data indicating that the previously received utterance was properly identified, properly rejected, falsely identified, or falsely rejected; and

in response to receiving the data indicating that the previously received utterance was properly identified, properly rejected, falsely identified, or falsely rejected, using audio data corresponding to the previously received utterance to perform voice-based speaker identification on subsequently received utterances.

2. The method of claim 1, comprising:

after receiving the data identifying the user, verifying an identification of the user using a technique other than voice-based speaker identification.

3. The method of claim 1, comprising:

prompting the user to verify their identity using a technique other than voice-based speaker identification.

4. The method of claim 3, wherein the technique other than voice-based speaker identification comprises prompting the user for a passcode.

5. The method of claim 1, wherein the previously received utterance and the subsequently received utterance include a predefined hotword.

6. The method of claim 1, wherein the audio data corresponding to the previously received utterance and audio data corresponding to the subsequently received utterance include an amount of background noise within a same background noise range.

7. The method of claim 6, wherein the amount of background noise is measured prior to receipt of the previously received utterance and the subsequently received utterance.

8. The method of claim 1, wherein prompting the user to confirm whether a previously received utterance was properly identified, properly rejected, falsely identified, or falsely rejected using voice-based speaker identification comprises:

providing, for display, data indicating a date and time when the previously received utterance was received.

9. The method of claim 1, wherein the audio data corresponding to the previously received utterance and audio data corresponding to the subsequently received utterance have a loudness within a same loudness range.

10. The method of claim 1, wherein the audio data corresponding to the previously received utterance and audio data corresponding to the subsequently received utterance have a signal-to-noise ratio within a same signal-to-noise ratio range.

11. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising;

receiving, by a computing device that uses voice-based speaker identification, data identifying a user;

prompting the user to confirm whether a previously received utterance was properly identified, properly rejected, falsely identified, or falsely rejected using voice-based speaker identification;

receiving, from the user, data indicating that the previously received utterance was properly identified, properly rejected, falsely identified, or falsely rejected; and

in response to receiving the data indicating that the previously received utterance was properly identified, properly rejected, falsely identified, or falsely rejected, using audio data corresponding to the previously received utterance to perform voice-based speaker identification on subsequently received utterances.

12. The system of claim 11, wherein the operations further comprise:

after receiving the data identifying the user, verifying an identification of the user using a technique other than voice-based speaker identification.

13. The system of claim 11, wherein the operations further comprise:

prompting the user to verify their identity using a technique other than voice-based speaker identification.

14. The system of claim 11, wherein the previously received utterance and the subsequently received utterance include a predefined hotword.

15. The system of claim 11, wherein the audio data corresponding to the previously received utterance and audio data corresponding to the subsequently received utterance include an amount of background noise within a same background noise range.

16. The system of claim 15, wherein the amount of background noise is measured prior to receipt of the previously received utterance and the subsequently received utterance.

17. The system of claim 11, wherein prompting the user to confirm whether a previously received utterance was properly identified, properly rejected, falsely identified, or falsely rejected using voice-based speaker identification comprises:

providing, for display, data indicating a date and time when the previously received utterance was received.

18. The system of claim 11, wherein the audio data corresponding to the previously received utterance and audio data corresponding to the subsequently received utterance have a loudness within a same loudness range.

19. The system of claim 11, wherein the audio data corresponding to the previously received utterance and audio data corresponding to the subsequently received utterance have a signal-to-noise ratio within a same signal-to-noise ratio range.

20. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:

receiving, by a computing device that uses voice-based speaker identification, data identifying a user;

prompting the user to confirm whether a previously received utterance was properly identified, properly rejected, falsely identified, or falsely rejected using voice-based speaker identification;

receiving, from the user, data indicating that the previously received utterance was properly identified, properly rejected, falsely identified, or falsely rejected; and

in response to receiving the data indicating that the previously received utterance was properly identified, properly rejected, falsely identified, or falsely rejected, using audio data corresponding to the previously received utterance to perform voice-based speaker identification on subsequently received utterances.