Information Processing Method And Electronic Apparatus

• US 20140136215A1
• Filed: 11/13/2013
• Published: 05/15/2014
• Est. Priority Date: 11/13/2012
• Status: Active Grant

• Current Assignee: Beijing Lenovo Software Ltd (Lenovo Group Ltd.), Lenovo Company Limited (Lenovo Group Ltd.)
• Original Assignee: Beijing Lenovo Software Ltd (Lenovo Group Ltd.), Lenovo Company Limited (Lenovo Group Ltd.)
• Inventors: Dai, Haisheng, Wang, Qianying, Wang, Hao, Wang, Tianshu, Li, Xiangyang, Fan, Lifeng
• Granted Patent: US 9,959,865 B2
• Time in Patent Office: Days
• US Class Current: 704/275
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

The present invention provides information processing method and electronic apparatus. The method is applied in an electronic apparatus having voice recognition service, and the method includes: obtaining first voice information; recognizing the first voice information by a first recognition model to obtain a first recognition result; deciding whether the first recognition result conforms to a first preset condition; recognizing the first voice information by a second recognition model different from the first recognition model to obtain a second recognition result when the first recognition result conforms to the first preset condition; and controlling the electronic apparatus to execute a corresponding control instruction based on the second recognition result.

Claims

1. An information processing method applied in an electronic apparatus having a voice recognition service, the method comprising:

obtaining first voice information;

recognizing the first voice information by a first recognition model to obtain a first recognition result;

deciding whether the first recognition result conforms to a first preset condition;

recognizing the first voice information by a second recognition model different from the first recognition model to obtain a second recognition result when the first recognition result conforms to the first preset condition;

controlling the electronic apparatus to execute a corresponding control instruction based on the second recognition result.

2. The method according to claim 1, wherein, when the first recognition result does not conform to the first preset condition, the method further comprises discarding the first voice information.

3. The method according to claim 1, wherein, before recognizing the first voice information by the first recognition model, the method further comprises:

deciding whether the first voice information satisfies a second preset condition;

discarding the first voice information when the first voice information does not satisfy the second preset condition; and

executing a step of recognizing the first voice information by the first recognition model when the first voice information satisfies the second preset condition.

4. The method according to claim 1, wherein, recognizing the first voice information by the first recognition model to obtain the first recognition result comprises recognizing whether the user corresponding to the first voice information is a predetermined user to obtain the first recognition result;

wherein, when the user corresponding to the first voice information is not the predetermined user, it represents that the first voice information does not satisfy the first preset condition, and when the user corresponding to the first voice information is the predetermined user, it represents that the first voice information satisfies the first preset condition.

5. The method according to claim 1, wherein, obtaining the first voice information comprises carrying out end-point detection to the first voice information to obtain the detected first voice information.

6. The method according to claim 1, wherein, when the first recognition result conforms to the first preset condition, recognizing the first voice information by the second recognition model different from the first recognition model to obtain the second recognition result comprises:

recognizing the first voice information by the second recognition model to obtain a third recognition result; and

,

obtaining the second recognition result based on the first recognition result and the third recognition result.

7. The method according to claim 1, wherein, the voice recognition service is in an inactive state, and when the control instruction corresponding to the second recognition result is a wakening instruction, controlling the electronic apparatus to execute a corresponding control instruction based on the second recognition result comprises executing the wakening instruction to waken the voice recognition service.

8. The method according to claim 1, wherein the electronic apparatus includes at least two different voice recognition modules, and the method further includes:

obtaining that the current mode of the electronic apparatus is a first usage mode;

determining the voice recognition module corresponding to the first usage mode is a first voice recognition module in the at least two different voice recognition modules based on correspondence relationship between the usage mode and the voice recognition module;

recognizing the first voice information by the second recognition model different from the first recognition model to obtain the second recognition result when the first recognition result conforms to the first preset condition includes recognizing the first voice information using the second recognition model different from the first recognition model by the first voice recognition module to obtain the second recognition result when the first recognition result conforms to the first preset condition.

9. The method according to claim 8, wherein, before obtaining that the current mode of the electronic apparatus is the first usage mode, the method further includes:

detecting to obtain second voice information when the electronic apparatus is in a second usage mode;

recognizing the second voice information by a second voice recognition module corresponding to the second usage mode in the at least two different voice recognition modules, to determine whether there is a first control instruction controlling the electronic apparatus to enter the first usage mode; and

controlling the electronic apparatus to enter the first usage mode when there is the first control instruction.

10. The method according to claim 9, wherein, after determining whether there is the first control instruction for controlling the electronic apparatus to enter the first usage mode, the method further comprises recording third voice information input by the user between a first moment of generating the first control instruction and a second moment of switching to the first usage mode when there is the first control instruction.

11. The method according to claim 10, wherein, after recording the third voice information input by the user when the electronic apparatus switches to the first usage mode from the second usage mode, the method further comprises storing the third voice information in a memory of the electronic apparatus.

12. The method according to claim 11, wherein, obtaining the first voice information comprises:

reading the third voice information from the memory; and

,

acquiring first sub-voice information that has not been carried out voice recognition by the second voice recognition module in the third voice information, the first sub-voice information being the first voice information.

13. The method according to claim 8, wherein, after controlling the electronic apparatus to enter the first usage mode, the method further comprises:

deciding whether there is a second control instruction for enabling the first voice recognition module; and

executing a step of determining that the voice recognition module corresponding to the first usage mode is the first voice recognition module based on the correspondence relationship between the usage mode and the voice recognition module after deciding there is the second control instruction.

14. The method according to claim 8, wherein, determining that the voice recognition module corresponding to the first usage mode is the first voice recognition module based on the correspondence relationship between the usage mode and the voice recognition module comprises:

determining that the voice recognition module corresponding to the first usage mode is a software voice recognition module based on the correspondence relationship between the usage mode and the voice recognition module when the first usage mode is a usage mode whose power consumption is larger than a first preset threshold; and

,

determining that the voice recognition module corresponding to the first usage mode is a hardware voice recognition module when the first usage mode is a usage mode whose power consumption is less than the first preset threshold.

15. The method according to claim 1, which is applied in an electronic apparatus, wherein, before obtaining the first voice information, the method further comprises:

detecting to obtain fourth voice information when the electronic apparatus is in a second usage mode, the second usage mode being a corresponding usage mode when the electronic apparatus is in a power saving status;

recognizing the fourth voice information by a second voice recognition module corresponding to the second usage mode, to determine whether there is a first control instruction for controlling the electronic apparatus to enter the first usage mode; and

controlling the electronic apparatus to enter the first usage mode when there is the first control instruction, wherein the first usage mode is a corresponding usage mode when the electronic apparatus is in a normal working status.

16. The method according to claim 15, wherein, acquiring the first voice information comprises:

determining the first voice recognition module corresponding to the first usage mode;

acquiring the first voice information; and

recognizing the first voice information by the second recognition model different from the first recognition model to obtain the second recognition result when the first recognition result conforms to the first preset condition comprising;

recognizing the first voice information using the second recognition model different from the first recognition model by the first voice recognition module to obtain the second recognition result when the first recognition result conforms to the first preset condition.

17. The method according to claim 1, wherein:

the electronic apparatus includes a first processing unit whose power consumption is a first power consumption and a second processing unit connected to the first processing unit, the power consumption of the second processing unit being a second power consumption which is larger than the first power consumption, and the second processing unit having voice recognition service,

obtaining the first voice information comprises;

the first processing unit obtaining the first voice information of a first user when the second processing unit is in a disabled status;

recognizing the first voice information by the first recognition model to obtain the first recognition result comprises;

the first processing unit recognizing the first voice information by the first recognition model to obtain the first recognition result;

deciding whether the first recognition result conforms to the first preset condition comprise;

the first processing unit deciding whether the first recognition result conforms to the first preset condition;

recognizing the first voice information by the second recognition model different from the first recognition model to obtain the second recognition result when the first recognition result conforms to the first preset condition comprises;

the first processing unit recognizing the first voice information by the second recognition model different from the first recognition model to obtain the second recognition result when the first recognition result conforms to the first preset condition;

controlling the electronic apparatus to execute a corresponding control instruction based on the second recognition result comprises;

the first processing unit generating a trigger instruction based on the second recognition result and sending the trigger instruction to the second processing unit to make the second processing unit be in an enabled status;

the second processing unit obtaining the second voice information of the first user;

the second processing unit generating a first acoustic control instruction corresponding to the second voice information based on correspondence relationship between the voice information and the acoustic control instruction.

18. The method according to claim 17, wherein, the second processing unit obtaining the second voice information of the first user comprises:

deciding whether there is content information conforming to the second preset condition in the first voice information; and

when there is no content information conforming to the second preset condition in the first voice information, the second processing unit obtaining the second voice information different from the first voice information.

19. An information processing method applied in an electronic apparatus including at least two different voice recognition modules, the method comprising:

obtaining first voice information to be recognized;

obtaining that the current mode of the electronic apparatus is a first usage mode;

determining that the voice recognition module corresponding to the first usage mode is a first voice recognition module in the at least two different voice recognition modules based on correspondence relationship between the usage mode and the voice recognition module; and

recognizing the first voice information to be recognized by the first voice recognition module to obtain a recognition result.

20. An information processing method applied in an electronic apparatus including a first processing unit whose power consumption is a first power consumption and a second processing unit connected to the first processing unit, wherein, the power consumption of the second processing unit is a second power consumption which is larger than the first power consumption, the method comprising:

when the second processing unit is in a disabled status, the first processing unit obtaining first sound information of a first user;

the first processing unit deciding whether the first sound information satisfies a first preset condition;

the first processing unit generating a trigger instruction and sending the trigger instruction to the second processing unit to make the second processing unit be in an enabled status when the first sound information satisfies the first preset condition;

the second processing unit obtaining second sound information of the first user;

the second processing unit generating a first acoustic control instruction corresponding to the second sound information based on correspondence relationship between the sound information and the acoustic control instruction.