Segment-based speaker verification using dynamically generated phrases
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for verifying an identity of a user. The methods, systems, and apparatus include actions of receiving a request for a verification phrase for verifying an identity of a user. Additional actions include, in response to receiving the request for the verification phrase for verifying the identity of the user, identifying subwords to be included in the verification phrase and in response to identifying the subwords to be included in the verification phrase, obtaining a candidate phrase that includes at least some of the identified subwords as the verification phrase. Further actions include providing the verification phrase as a response to the request for the verification phrase for verifying the identity of the user.
1. A computer-implemented method when executed on data processing hardware causes the data processing hardware to perform operations comprising:
receiving a command input by a particular user;
in response to receiving the command, generating a verification phrase for the particular user, the verification phrase comprising at least one subword obtained from stored enrollment audio data representing the particular user speaking the subword;
providing, as output from a user interface, a message informing the particular user to speak the verification phrase to verify the identity of the particular user;
receiving verification audio data representing an unverified user speaking the verification phrase;
determining whether the unverified user speaking the verification phrase comprises the particular user based on the stored enrollment audio data and the verification audio data; and
in response to determining that the unverified user speaking the verification phrase comprises the particular user, verifying an identity of the unverified user as the particular user.
2. The computer-implemented method of claim 1 , wherein providing the message as output from the user interface comprises displaying the message as output on a display in communication with the data processing hardware.
3. The computer-implemented method of claim 1 , wherein providing the message as output from the user interface comprises outputting the message as synthesized speech.
4. The computer-implemented method of claim 1 , wherein the operations further comprise, after verifying the identity of the unverified user as the particular user, displaying a welcome interface on a display in communication with the data processing hardware.
5. The computer-implemented method of claim 1 , wherein the operations further comprise:
receiving the enrollment audio data representing the particular user speaking the at least one subword; and
storing the received enrollment audio data in memory hardware in communication with the data processing hardware.
6. The computer-implemented method of claim 5 , wherein the operations further comprises determining that the received enrollment audio data contains a minimum number of subwords spoken by the particular user a minimum number of times.
7. The computer-implemented method of claim 6 , wherein the subwords comprise phonemes.
8. The computer-implemented method of claim 5 , wherein receiving the enrollment audio data continues until enrollment audio data has been received that satisfies an utterance quality threshold.
9. The computer-implemented method of claim 1 , wherein determining whether the unverified user speaking the verification phrase comprises the particular user comprises:
determining whether the verification audio data matches the stored enrollment audio data; and
in response to determining that the verification audio data matches the stored enrollment audio data, classifying the unverified user as the particular user.
10. A system:
data processing hardware: and
memory hardware in communication with the data processing hardware and storing instructions that when executed on the data processing hardware cause the data processing hardware to perform operations comprising:
receiving a command input by a particular user;
in response to receiving the command, generating a verification phrase for the particular user, the verification phrase comprising at least one subword obtained from stored enrollment audio data representing the particular user speaking the subword;
providing, as output from a user interface, a message informing the particular user to speak the verification phrase to verify the identity of the particular user;
receiving verification audio data representing an unverified user speaking the verification phrase;
determining whether the unverified user speaking the verification phrase comprises the particular user based on the stored enrollment audio data and the verification audio data; and
in response to determining that the unverified user speaking the verification phrase comprises the particular user, verifying an identity of the unverified user as the particular user.
11. The system of claim 10 , wherein determining whether the unverified user speaking the verification phrase comprises the particular user comprises:
determining whether the verification audio data matches the stored enrollment audio data; and
in response to determining that the verification audio data matches the stored enrollment audio data, classifying the unverified user as the particular user.
12. The system of claim 10 , wherein providing the message as output from the user interface comprises displaying the message as output on a display in communication with the data processing hardware.
13. The system of claim 10 , wherein providing the message as output from the user interface comprises outputting the message as synthesized speech.
14. The system of claim 10 , wherein the operations further comprise, after verifying the identity of the unverified user as the particular user, displaying a welcome interface on a display in communication with the data processing hardware.
15. The system of claim 10 , wherein the operations further comprise:
receiving the enrollment audio data representing the particular user speaking the at least one subword; and
storing the received enrollment audio data in memory hardware in communication with the data processing hardware.
16. The system of claim 15 , wherein the operations further comprises determining that the received enrollment audio data contains a minimum number of subwords spoken by the particular user a minimum number of times.
17. The system of claim 16 , wherein the subwords comprise phonemes.
18. The system of claim 16 , wherein receiving the enrollment audio data continues until enrollment audio data has been received that satisfies an utterance quality threshold.