CN-113870865-B - Voiceprint feature updating method and device, electronic equipment and storage medium

Abstract

The embodiment of the application provides a voiceprint feature updating method, a voiceprint feature updating device, electronic equipment and a storage medium, which relate to the technical field of biological recognition and comprise the steps of obtaining voice data of a user to be verified, extracting voiceprint features of the user to be verified according to the obtained voice data, obtaining other biological data of the user to be verified, extracting other biological features of the user to be verified according to the obtained other biological data, calculating first similarity between the extracted voiceprint features and registered voiceprint features of a registered user, and calculating second similarity between the extracted other biological features and the other biological features of the registered user, wherein the voiceprint features of the registered user are updated by using the extracted voiceprint features under the condition that the first similarity is larger than a voiceprint fault tolerance threshold and the second similarity is larger than an identity characterization threshold. By applying the voiceprint feature updating scheme provided by the embodiment of the application, the reliability of identity verification can be improved.

Inventors

• ZHOU JIANFENG

Assignees

• 杭州海康威视数字技术股份有限公司

Dates

Publication Date

20260508

Application Date

20211012

Claims (9)

1. 1. A method of voiceprint feature updating, the method comprising: Obtaining voice data of a user to be verified, extracting voiceprint characteristics of the user to be verified according to the obtained voice data, obtaining other biological data of the user to be verified, and extracting other biological characteristics of the user to be verified according to the obtained other biological data, wherein the other biological data are biological data which support identity verification except the voice data; calculating a first similarity between the extracted voiceprint features and voiceprint features of a registered user and calculating a second similarity between the extracted other biometric features and other biometric features of the registered user; Updating the voiceprint features of the registered user by using the extracted voiceprint features under the condition that the first similarity is larger than a voiceprint fault tolerance threshold and smaller than a voiceprint verification threshold and the second similarity is larger than an identity characterization threshold, wherein the voiceprint fault tolerance threshold is smaller than the voiceprint verification threshold, the identity characterization threshold is larger than other feature verification thresholds, and the voiceprint verification threshold and the other feature verification thresholds are thresholds for identity verification based on voiceprint features and other biological features; determining that the user to be authenticated and the registered user are the same user when the first similarity is greater than the voiceprint authentication threshold and the second similarity is greater than the other feature authentication threshold, and/or And under the condition that the first similarity is larger than a voiceprint fault tolerance threshold and the second similarity is larger than an identity characterization threshold, determining that the user to be verified and the registered user are the same user.
2. 2. The method of claim 1, wherein updating the voiceprint features of the registered user with the extracted voiceprint features if the first similarity is greater than a voiceprint fault tolerance threshold and the second similarity is greater than an identity characterization threshold comprises: determining a voice quality of the obtained voice data; And under the condition that the determined voice quality meets the preset voice quality condition, the first similarity is larger than a voiceprint fault tolerance threshold value, and the second similarity is larger than an identity characterization threshold value, updating the voiceprint characteristics of the registered user by using the extracted voiceprint characteristics.
3. 3. The method of claim 2, wherein updating the voiceprint features of the registered user with the extracted voiceprint features if the determined voice quality meets a preset voice quality condition, the first similarity is greater than a voiceprint fault tolerance threshold, and the second similarity is greater than an identity characterization threshold, comprises: determining a data quality of the obtained other biological data; And under the condition that the determined voice quality meets the preset voice quality condition, the determined data quality meets the preset data quality condition, the first similarity is larger than a voiceprint fault tolerance threshold value, and the second similarity is larger than an identity characterization threshold value, updating the voiceprint characteristics of the registered user by utilizing the extracted voiceprint characteristics.
4. 4. The method of claim 2, wherein said determining the voice quality of the obtained voice data comprises: Determining the quality of the voiceprint environment reflecting the noisy degree of the speaking environment of the user to be verified according to the obtained voice data, and/or And carrying out semantic recognition on the obtained voice data, and determining the voiceprint content quality reflecting the speaking content of the user to be verified based on the recognition result.
5. 5. The method of claim 1, wherein updating the voiceprint features of the registered user with the extracted voiceprint features if the first similarity is greater than a voiceprint fault tolerance threshold and the second similarity is greater than an identity characterization threshold comprises: storing the extracted voiceprint features as target features for updating the voiceprint features of the registered user if the first similarity is greater than a voiceprint fault tolerance threshold and the second similarity is greater than an identity characterization threshold; and under the condition that the number of the stored target features reaches a preset number, updating the voiceprint features of the registered user by utilizing the stored target features.
6. 6. The method of claim 5, wherein storing the extracted voiceprint features as target features for updating voiceprint features of the registered user if the first similarity is greater than a voiceprint fault tolerance threshold and the second similarity is greater than an identity characterization threshold, comprises: And under the condition that the first similarity is larger than a voiceprint fault tolerance threshold value, the second similarity is larger than an identity characterization threshold value and a time interval exceeds a preset duration, storing the extracted voiceprint features as target features for updating the voiceprint features of the registered user, wherein the time interval is the interval between the last time of storing the target features and the current time.
7. 7. A voiceprint feature updating apparatus, the apparatus comprising: The voice print characteristic obtaining module is used for obtaining voice data of the user to be verified and extracting voice print characteristics of the user to be verified according to the obtained voice data; The other characteristics obtaining module is used for obtaining other biological data of the user to be verified, and extracting other biological characteristics of the user to be verified according to the obtained other biological data, wherein the other biological data are biological data which are except voice data and support identity verification; A first similarity calculation module, configured to calculate a first similarity between the extracted voiceprint feature and a voiceprint feature of a registered user, and calculate a second similarity between the extracted other biometric feature and the other biometric feature of the registered user; The feature updating module is used for updating the voiceprint feature of the registered user by utilizing the extracted voiceprint feature under the condition that the first similarity is larger than a voiceprint fault tolerance threshold and smaller than a voiceprint verification threshold and the second similarity is larger than an identity characterization threshold, wherein the voiceprint fault tolerance threshold is smaller than the voiceprint verification threshold, the identity characterization threshold is larger than other feature verification thresholds, and the voiceprint verification threshold and the other feature verification thresholds are thresholds used for carrying out identity verification based on the voiceprint feature and other biological features; The device further comprises an identity verification module which is used for determining that the user to be verified and the registered user are the same user when the first similarity is larger than the voiceprint verification threshold value and the second similarity is larger than the other characteristic verification threshold value, and/or determining that the user to be verified and the registered user are the same user when the first similarity is larger than the voiceprint fault tolerance threshold value and the second similarity is larger than the identity characterization threshold value.
8. 8. The electronic equipment is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus; A memory for storing a computer program; A processor for carrying out the method steps of any one of claims 1-6 when executing a program stored on a memory.
9. 9. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored therein a computer program which, when executed by a processor, implements the method steps of any of claims 1-6.

Description

Voiceprint feature updating method and device, electronic equipment and storage medium Technical Field The present application relates to the field of biological identification technologies, and in particular, to a voiceprint feature updating method, a voiceprint feature updating device, an electronic device, and a storage medium. Background In biometric identification, it is often necessary to authenticate a user using its characteristics. For example, the user may be authenticated using the facial features of the user. In order to ensure the reliability of the authentication, the authentication is generally performed by using multi-mode features, for example, voiceprint features and face features. In the related technology, voiceprint features and human face features of a user are generally obtained and stored in a registration stage, the voiceprint features and the human face features of the user to be verified are obtained in a verification stage, the voiceprint features and the human face features are matched with the voiceprint features and the human face features stored in the registration stage, and under the condition that the voiceprint features and the human face features are successfully matched, the user to be verified is considered to pass verification. Although the scheme can realize the authentication by adopting the multi-mode characteristics, the sounding organ and the sounding mode of the user can be changed along with the change of time, so that the voiceprint characteristics of the user can be changed, and the voiceprint characteristics extracted in the authentication stage can be difficult to match with the voiceprint characteristics in the registration stage, thereby reducing the reliability of the authentication. Disclosure of Invention The embodiment of the application aims to provide a voiceprint feature updating method, a voiceprint feature updating device, electronic equipment and a storage medium, so as to improve the reliability of identity verification. The specific technical scheme is as follows: In a first aspect, an embodiment of the present application provides a method for updating a voiceprint feature, where the method includes: Obtaining voice data of a user to be verified, extracting voiceprint characteristics of the user to be verified according to the obtained voice data, obtaining other biological data of the user to be verified, and extracting other biological characteristics of the user to be verified according to the obtained other biological data, wherein the other biological data are biological data which support identity verification except the voice data; calculating a first similarity between the extracted voiceprint features and voiceprint features of a registered user and calculating a second similarity between the extracted other biometric features and other biometric features of the registered user; and under the condition that the first similarity is larger than a voiceprint fault tolerance threshold and the second similarity is larger than an identity characterization threshold, updating voiceprint features of the registered user by using the extracted voiceprint features. In one embodiment of the application, the voiceprint fault tolerance threshold is smaller than the voiceprint verification threshold, the identity characterization threshold is larger than other characteristic verification thresholds, and the voiceprint verification threshold and other characteristic verification thresholds are thresholds for identity verification based on voiceprint characteristics and other biological characteristics; The method further comprises the steps of: determining that the user to be authenticated and the registered user are the same user when the first similarity is greater than the voiceprint authentication threshold and the second similarity is greater than the other feature authentication threshold, and/or And under the condition that the first similarity is larger than a voiceprint fault tolerance threshold and the second similarity is larger than an identity characterization threshold, determining that the user to be verified and the registered user are the same user. In one embodiment of the present application, the updating the voiceprint feature of the registered user using the extracted voiceprint feature if the first similarity is greater than a voiceprint fault tolerance threshold and the second similarity is greater than an identity characterization threshold includes: and under the condition that the first similarity is larger than a voiceprint fault tolerance threshold and smaller than the voiceprint verification threshold and the second similarity is larger than an identity characterization threshold, updating voiceprint features of the registered user by using the extracted voiceprint features. In one embodiment of the present application, the updating the voiceprint feature of the registered user using the extracted voiceprint feature if t