Voiceprint recognition model construction
Technologies related to voiceprint recognition model construction are disclosed. In an implementation, a first voice input from a user is received. One or more predetermined keywords from the first voice input are detected. One or more voice segments corresponding to the one or more predetermined keywords are recorded. The voiceprint recognition model is trained based on the one or more voice segments. A second voice input is received from a user, and the user's identity is verified based on the second voice input using the voiceprint recognition model.
1. A computer-implemented method, comprising:
receiving a first voice input from a user;
obtaining a security requirement level of a system for training a voiceprint recognition model for the user;
determining a minimum number of required keywords for training the voiceprint recognition model for the user, the minimum number of required keywords being based on the security requirement level of the system;
detecting one or more predetermined keywords from the first voice input, wherein the one or more predetermined keywords include at least the minimum number of required keywords for the security requirement level;
recording one or more voice segments corresponding to the one or more predetermined keywords;
training the voiceprint recognition model based on the one or more voice segments corresponding to the minimum number of required keywords for the security requirement level;
after training the voiceprint recognition model, receiving, from the user, second voice input;
determining that the second voice input comprises a particular keyword that was not used to train the voiceprint recognition model;
in response:
capturing the second voice input having a representation of the particular keyword dictated by the user; and
updating the voiceprint recognition model using the second voice input having the representation of the particular keyword dictated by the user that was not used to train the voiceprint recognition model;
receiving, from the user, third voice input that comprises at least the minimum number of required keywords for the security requirement level; and
verifying an identity of the user based on the third voice input using the trained voiceprint recognition model.
2. The computer-implemented method of claim 1 , further comprising detecting one or more predetermined keywords from the third voice input.
3. The computer-implemented method of claim 2 , further comprising recording one or more voice segments corresponding to the one or more predetermined keywords from the third voice input.
4. The computer-implemented method of claim 3 , further comprising updating the voiceprint recognition model based on the one or more voice segments corresponding to the one or more predetermined keywords from the third voice input.
5. The computer-implemented method of claim 1 , wherein detecting the one or more predetermined keywords from the first voice input is based on an acoustic model and the one or more voice segments include one or more acoustic features same as the one or more predetermined keywords.
6. The computer-implemented method of claim 1 , wherein detecting the one or more predetermined keywords from the first voice input is performed based on voice recognition.
7. A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:
receiving a first voice input from a user;
obtaining a security requirement level of a system for training a voiceprint recognition model for the user;
determining a minimum number of required keywords for training the voiceprint recognition model for the user, the minimum number of required keywords being based on the security requirement level of the system;
detecting one or more predetermined keywords from the first voice input, wherein the one or more predetermined keywords include at least the minimum number of required keywords for the security requirement level;
recording one or more voice segments corresponding to the one or more predetermined keywords;
training the voiceprint recognition model based on the one or more voice segments corresponding to the minimum number of required keywords for the security requirement level;
after training the voiceprint recognition model, receiving, from the user, second voice input;
determining that the second voice input comprises a particular keyword that was not used to train the voiceprint recognition model;
in response:
capturing the second voice input having a representation of the particular keyword dictated by the user; and
updating the voiceprint recognition model using the second voice input having the representation of the particular keyword dictated by the user that was not used to train the voiceprint recognition model;
receiving, from the user, third voice input that comprises at least the minimum number of required keywords for the security requirement level; and
verifying an identity of the user based on the third voice input using the trained voiceprint recognition model.
8. The non-transitory, computer-readable medium of claim 7 , further comprising detecting one or more predetermined keywords from the third voice input.
9. The non-transitory, computer-readable medium of claim 8 , further comprising recording one or more voice segments corresponding to the one or more predetermined keywords from the third voice input.
10. The non-transitory, computer-readable medium of claim 9 , further comprising updating the voiceprint recognition model based on the one or more voice segments corresponding to the one or more predetermined keywords from the third voice input.
11. The non-transitory, computer-readable medium of claim 7 , wherein detecting the one or more predetermined keywords from the first voice input is based on an acoustic model and the one or more voice segments include one or more acoustic features same as the one or more predetermined keywords.
12. The non-transitory, computer-readable medium of claim 7 , wherein detecting the one or more predetermined keywords from the first voice input is performed based on voice recognition.
13. A computer-implemented system, comprising:
one or more computers; and
one or more computer memory devices interoperably coupled with the one or more computers and having tangible, non-transitory, machine-readable media storing one or more instructions that, when executed by the one or more computers, perform one or more operations comprising:
receiving a first voice input from a user;
obtaining a security requirement level of a system for training a voiceprint recognition model for the user;
determining a minimum number of required keywords for training the voiceprint recognition model for the user, the minimum number of required keywords being based on the security requirement level of the system;
detecting one or more predetermined keywords from the first voice input, wherein the one or more predetermined keywords include at least the minimum number of required keywords for the security requirement level;
recording one or more voice segments corresponding to the one or more predetermined keywords;
training the voiceprint recognition model based on the one or more voice segments corresponding to the minimum number of required keywords for the security requirement level;
after training the voiceprint recognition model, receiving, from the user, second voice input;
determining that the second voice input comprises a particular keyword that was not used to train the voiceprint recognition model;
in response:
capturing the second voice input having a representation of the particular keyword dictated by the user; and
updating the voiceprint recognition model using the second voice input having the representation of the particular keyword dictated by the user that was not used to train the voiceprint recognition model;
receiving, from the user, third voice input that comprises at least the minimum number of required keywords for the security requirement level; and
verifying an identity of the user based on the third voice input using the trained voiceprint recognition model.
14. The computer-implemented system of claim 13 , further comprising detecting one or more predetermined keywords from the third voice input.
15. The computer-implemented system of claim 14 , further comprising recording one or more voice segments corresponding to the one or more predetermined keywords from the third voice input.
16. The computer-implemented system of claim 15 , further comprising updating the voiceprint recognition model based on the one or more voice segments corresponding to the one or more predetermined keywords from the third voice input.
17. The computer-implemented system of claim 13 , further comprising determining the one or more predetermined keywords from the third voice input.
18. The computer-implemented system of claim 13 , wherein detecting the one or more predetermined keywords from the first voice input is based on an acoustic model and the one or more voice segments include one or more acoustic features same as the one or more predetermined keywords.