Voice activation method, apparatus, electronic device, and storage medium

• US 11,069,343 B2
• Filed: 05/22/2019
• Issued: 07/20/2021
• Est. Priority Date: 02/16/2017
• Status: Active Grant

• Current Assignee: Tencent Technology Company Limited (Tencent Holdings Limited)
• Original Assignee: Tencent Technology Company Limited (Tencent Holdings Limited)
• Inventors: Xiang, Qianbiao, Chen, Menglin
• Primary Examiner(s): Islam, Mohammad K
• Application Number:
  US16/419,924
• Publication Number:
  US 20190272820A1
• Time in Patent Office: 790 Days
• CPC Class Codes: B60K 2360/148 Instrument input by voice
• CPC Class Codes: B60K 35/00 Instruments specially adapted for vehicles; Arrangement of instruments in or on vehicles
• CPC Class Codes: B60K 35/10 Input arrangements, i.e. from user to vehicle, associated with vehicle functions or specially adapted therefor
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 2025/783 based on threshold decision
• CPC Class Codes: G10L 25/51 for comparison or discrimination
• CPC Class Codes: G10L 25/84 for discriminating voice from noise

Abstract

Embodiments of the present disclosure disclose a voice activation method, an apparatus, an electronic device, and a storage medium thereof. The voice activation method includes: capturing a voice signal in a vehicle; calculating an acoustic score of an activation keyword extracted from the voice signal, the acoustic score being used for indicating authenticity of the activation keyword, the authenticity of the activation keyword being a probability that the activation keyword is used for waking up the electronic device to perform a voice activation operation, and a magnitude of the acoustic score being positively correlated to a magnitude of the probability; determining a voice activation threshold according to a noise level in the vehicle, a magnitude of the voice activation threshold being negatively correlated to a magnitude of the noise; and performing the voice activation operation when the acoustic score is greater than the voice activation threshold.

Claims

1. A voice activation method applied to an electronic device in a vehicle, comprising:

capturing a voice signal in a vehicle, the voice signal including a first voice signal captured in a first period from a first timestamp to a second timestamp and a second voice signal captured in a second period from a third timestamp to the first timestamp;

calculating an acoustic score of an activation keyword extracted from the first voice signal, the acoustic score being used for indicating authenticity of the activation keyword, the authenticity of the activation keyword being a probability that the activation keyword is used for waking up the electronic device to perform a voice activation operation, and a magnitude of the acoustic score being positively correlated to a magnitude of the probability;

calculating a noise level in the vehicle by using the second voice signal;

determining a voice activation threshold according to the noise level in the vehicle, a magnitude of the voice activation threshold being negatively correlated to a magnitude of the noise;

comparing the acoustic score of the first voice signal with the voice activation threshold determined according to the noise level calculated from the second voice signal; and

performing the voice activation operation when the acoustic score is greater than the voice activation threshold.

2. The method according to claim 1, wherein before the determining a voice activation threshold according to the noise level in the vehicle, the method further comprises:

determining the noise level in the vehicle according to an average decibel value of the second voice signal.

3. The method according to claim 1, wherein before the determining a voice activation threshold according to the noise level in the vehicle, the method further comprises:

determining the noise level in the vehicle according to a vehicle velocity, the magnitude of the noise level is positively correlated to a velocity of the vehicle.

4. The method according to claim 1, wherein the determining a voice activation threshold according to noise in the vehicle comprises:

obtaining a noise interval and a voice activation threshold interval, each value in the noise interval corresponding to one value in the voice activation threshold interval; and

selecting the voice activation threshold value according to a value of the noise in the noise interval.

5. The method according to claim 2, wherein the determining a voice activation threshold according to noise in the vehicle comprises:

obtaining a noise interval and a voice activation threshold interval, each value in the noise interval corresponding to one value in the voice activation threshold interval; and

selecting the voice activation threshold value according to a value of the noise in the noise interval.

6. The method according to claim 3, wherein the determining a voice activation threshold according to noise in the vehicle comprises:

obtaining a noise interval and a voice activation threshold interval, each value in the noise interval corresponding to one value in the voice activation threshold interval; and

selecting the voice activation threshold value according to a value of the noise in the noise interval.

7. The method according to claim 1, wherein the electronic device in the vehicle is an electronic device mounted in the vehicle or a mobile terminal placed in the vehicle, and the electronic device mounted in the vehicle is a center console, an onboard device, or an intelligent rear view mirror.

17. The method according to claim 1, wherein capturing the voice signal in the vehicle comprises:

continuously capturing, by a microphone of the electronic device, the voice signal in the vehicle; and

sending, by the microphone at intervals of a predetermined time period, segments of the voice signal to a speech recognition module of the electronic device, each segment of the voice signal having the same time length as the predetermined time period;

wherein the first voice signal is one segment of the voice signal, and the second voice signal is one or more segments of the voice signal.

18. The method according to claim 1, wherein the voice activation threshold corresponding to the first voice signal is determined before the first voice signal is received.

19. The method according to claim 1, wherein:

a time length of the second period corresponding to the second voice signal is an integer multiple of a time length of the first period corresponding to the first voice signal.

8. An electronic device, the electronic device comprising:

one or more processors; and

a memory,

the memory storing one or more programs, the one or more programs being configured to be executed by the one or more processors, and the one or more programs comprising instructions for performing the following operations;

capturing a voice signal in a vehicle, the voice signal including a first voice signal captured in a first period from a first timestamp to a second timestamp and a second voice signal captured in a second period from a third timestamp to the first timestamp;

calculating an acoustic score of an activation keyword extracted from the first voice signal, the acoustic score being used for indicating authenticity of the activation keyword, the authenticity of the activation keyword being a probability that the activation keyword is used for waking up the electronic device to perform a voice activation operation, and a magnitude of the acoustic score being positively correlated to a magnitude of the probability;

calculating a noise level in the vehicle by using the second voice signal;

determining a voice activation threshold according to the noise level in the vehicle, a magnitude of the voice activation threshold being negatively correlated to a magnitude of the noise;

comparing the acoustic score of the first voice signal with the voice activation threshold determined according to the noise level calculated from the second voice signal; and

performing the voice activation operation when the acoustic score is greater than the voice activation threshold.

9. The electronic device according to claim 8, wherein the one or more programs further comprise instructions for performing the following operation:

determining the noise level in the vehicle according to an average decibel value of the second voice signal.

10. The electronic device according to claim 8, wherein the one or more programs further comprise an instruction used for performing the following operation:

determining the noise in the vehicle according to a vehicle velocity before the determining a voice activation threshold according to noise in the vehicle, wherein the magnitude of the noise is positively correlated to a magnitude of the vehicle velocity.

11. The electronic device according to claim 8, wherein the one or more programs further comprise an instruction used for performing the following operations:

obtaining a noise interval and a voice activation threshold interval, wherein each value in the noise interval corresponds to one value in the voice activation threshold interval; and

selecting the voice activation threshold interval, according to a value of the noise in the noise interval, a value corresponding to the value in the noise interval of the voice activation threshold.

12. The electronic device according to claim 9, wherein the one or more programs further comprise an instruction used for performing the following operations:

obtaining a noise interval and a voice activation threshold interval, wherein each value in the noise interval corresponds to one value of the voice activation threshold interval; and

selecting the voice activation threshold interval, according to a value of the noise in the noise interval, a value corresponding to the value in the noise interval of the voice activation threshold.

13. The electronic device according to claim 8, wherein the electronic device in the vehicle is an electronic device mounted in the vehicle or a mobile terminal placed in the vehicle, and the electronic device mounted in the vehicle is a center console, an onboard device, or an intelligent rear view mirror.

14. A non-transitory computer-readable storage medium storing computer program instructions executable by at least one processor to perform:

capturing a voice signal in a vehicle, the voice signal including a first voice signal captured in a first period from a first timestamp to a second timestamp and a second voice signal captured in a second period from a third timestamp to the first timestamp;

calculating an acoustic score of an activation keyword extracted from the voice signal, the acoustic score being used for indicating authenticity of the activation keyword, the authenticity of the activation keyword being a probability that the activation keyword is used for waking up the electronic device to perform a voice activation operation, and a magnitude of the acoustic score being positively correlated to a magnitude of the probability;

calculating a noise level in the vehicle by using the second voice signal;

determining a voice activation threshold according to the noise level in the vehicle, a magnitude of the voice activation threshold being negatively correlated to a magnitude of the noise;

comparing the acoustic score of the first voice signal with the voice activation threshold determined according to the noise level calculated from the second voice signal; and

performing the voice activation operation when the acoustic score is greater than the voice activation threshold.

15. The storage medium according to claim 14, wherein the computer program instructions further cause the at least one processor to perform:

determining the noise level in the vehicle according to an average decibel value of the second voice signal.

16. The storage medium according to claim 14, wherein the computer program instructions further cause the at least one processor to perform:

determining the noise in the vehicle according to a vehicle velocity before the determining a voice activation threshold according to noise in the vehicle, wherein the magnitude of the noise is positively correlated to a magnitude of the vehicle velocity.