Voiceprint security with messaging services

Patent metadata

Field	Value
Publication number	US-10715522-B2
Application number	US-201815885585-A
Country	US
Kind code	B2
Filing date	Jan 31, 2018
Priority date	Jan 31, 2018
Publication date	Jul 14, 2020
Grant date	Jul 14, 2020

Abstract

An online system authenticates a user through a voiceprint biometric verification process. When a user needs to be authenticated, the online system generates and provides a random phrase to the user. The online system receives an audio recording of the randomly generated phrase and retrieves a previously trained voiceprint model for the user. The online system analyzes the audio recording by applying the voiceprint model to determine whether the audio recording satisfies a first criteria of whether the voice in the audio recording belongs the user and a second criteria of whether the audio recording includes a vocalization of the randomly generated phrase. If the audio recording satisfies both criteria, the online system authenticates the user. Therefore, the user can be provided access to a new communication session in response to being authenticated.

First claim

What is claimed is:

1. A method comprising:
receiving, from a client device, a request for a new communication session;
generating a random phrase for a user of the client device for the new communication session;
providing, to the client device, the generated random phrase;
receiving an audio recording of the random phrase from the client device;
extracting a first set of features and a second set of features from the audio recording, the first set of features defining a voice of the user, the second set of features defining the random phrase that is vocalized in the audio recording;
applying the first set of features and the second set of features to a trained voiceprint model previously generated for the user to obtain an authentication score generated by a first portion of the trained voiceprint model and a predicted phrase generated by a second portion of the trained voiceprint model, the authentication score representing a similarity between the audio recording from the client device and prior audio recordings of the user, the predicted phrase representing a prediction of the vocalized random phrase in the audio recording; and
authorizing the client device to access the new communication session based on the obtained authentication score and predicted phrase.

2. The method of claim 1 , wherein generating the random phrase for the user of the client device comprises:
comparing the generated random phrase to one or more phrases previously generated for the user.

3. The method of claim 1 , wherein extracting the first set of features and the second set of features comprises:
extracting voice pattern features from the audio recording.

4. The method of claim 3 , wherein the extracted voice pattern features comprise one or more of a statistical measure of an amplitude of the audio recording, a statistical measure of a frequency of the audio recording, a cadence, a length of time for a particular sound, an accent, a lisp, a categorization of a phoneme, a cepstral coefficient, a perceptual linear predictive coefficient, and a filter-bank feature.

5. The method of claim 1 , wherein the new communication session is one of a new phone call, a new online chat, a new voice call, video call, or a new text message.

6. The method of claim 1 , wherein the trained voiceprint model is previously generated for the user by:
providing a training phrase, the training phase randomly generated for the user
obtaining a training audio recording of the user, the audio recording corresponding to the training phrase;
extracting voice pattern features from the training audio recording; and
training the voiceprint model based on the extracted voice pattern features from the training audio recording.

7. The method of claim 1 further comprising:
receiving, from a client device, a first level of authentication information; and
verifying the first level of authentication information,
wherein authorizing the client device to access the new communication session is further based on the verification of the first level of authentication information.

8. A non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to:
receive, from a client device, a request for a new communication session;
generate a random phrase for a user of the client device for the new communication session;
provide, to the client device, the generated random phrase;
receive an audio recording of the random phrase from the client device;
extract a first set of features and a second set of features from the audio recording, the first set of features defining a voice of the user, the second set of features defining the random phrase that is vocalized in the audio recording;
apply the first set of features and the second set of features to a trained voiceprint model previously generated for the user to obtain an authentication score generated by a first portion of the trained voiceprint model and a predicted phrase generated from a second portion of the trained voiceprint model, the authentication score representing a similarity between the audio recording from the client device and prior audio recordings of the user, the predicted phrase representing a prediction of the vocalized random phrase in the audio recording; and
authorize the client device to access the new communication session based on the obtained authentication score and predicted phrase.

9. The non-transitory computer readable medium of claim 8 , wherein the instructions that cause the processor to generate the random phrase for the user of the client device further comprises instructions that, when executed by the processor, cause the processor to:
compare the generated random phrase to one or more phrases previously generated for the user.

10. The non-transitory computer readable medium of claim 8 , wherein the instructions to extract the first set of features and the second set of features further comprise instructions that, when executed by the processor, cause the processor to:
extract voice pattern features from the audio recording.

11. The non-transitory computer readable medium of claim 10 , wherein the extracted voice pattern features comprise one or more of a statistical measure of an amplitude of the audio recording, a statistical measure of a frequency of the audio recording, a cadence, a length of time for a particular sound, an accent, a lisp, a categorization of a phoneme, a cepstral coefficient, a perceptual linear predictive coefficient, and a filter-bank feature.

12. The non-transitory computer readable medium of claim 8 , wherein the new communication session is one of a new phone call, a new online chat, a new voice call, video call, or a new text message.

13. The non-transitory computer readable medium of claim 8 , further comprising instructions that, when executed by a processor, cause the processor to:
receive, from a client device, a first level of authentication information; and
verify the first level of authentication information,
wherein the authorization of the client device to access the new communication session is further based on the verification of the first level of authentication information.

14. A method comprising:
receiving, from a service provider system, a request for a random phrase for authorization of a new communication session for a user of a client device;
generating a random phrase for the user of the client device for the new communication session;
transmitting the generated random phrase to the service provider system;
receiving an audio recording of the random phrase;
extracting a first set of features and a second set of features from the audio recording, the first set of features defining a voice of the user, the second set of features defining the random phrase that is vocalized in the audio recording;
applying the first set of features and the second set of features to a trained voiceprint model previously generated for the user to obtain an authentication score generated by a first portion of the trained voiceprint model and a predicted phrase generated by a second portion of the trained voiceprint model, the authentication score representing a similarity between the audio recording and prior audio recordings of the user, the predicted phrase representing a prediction of the vocalized random phrase in the audio recording; and
providing an authorization to the service provider system based on the obtained authentication score and predicted phrase, the authorization causing the service provider system to provide the user of the client device access to the new communication sessio

Assignees

• Salesforce Com Inc
• Salesforce Com

Inventors

• Lew Eugene

Classifications

• G10L17/00

  Speaker identification or verification techniques · CPC title

• H04L63/10

  for controlling access to devices or network resources · CPC title

• G06F21/32

  using biometric data, e.g. fingerprints, iris scans or voiceprints · CPC title

• H04L65/1069

  Session establishment or de-establishment · CPC title

• H04L63/0861 Primary

  using biometrical features, e.g. fingerprint, retina-scan (cryptographic mechanisms or cryptographic arrangements for entity authentication using biological data H04L9/3231) · CPC title