VOICE RECOGNITION METHOD, VOICE CONTROLLING METHOD, INFORMATION PROCESSING METHOD, AND ELECTRONIC APPARATUS

• US 20150142438A1
• Filed: 08/12/2014
• Published: 05/21/2015
• Est. Priority Date: 11/18/2013
• Status: Active Grant

• Current Assignee: Beijing Lenovo Software Ltd (Lenovo Group Ltd.), Lenovo (Beijing) Limited (Lenovo Group Ltd.)
• Original Assignee: Beijing Lenovo Software Ltd (Lenovo Group Ltd.), Lenovo (Beijing) Limited (Lenovo Group Ltd.)
• Inventors: Dai, Haisheng, Wang, Qianying, Chen, Shi
• Granted Patent: US 9,443,522 B2
• Time in Patent Office: Days
• US Class Current: 704/246
• CPC Class Codes: G10L 17/02 Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction
• CPC Class Codes: G10L 17/04 Training, enrolment or model building
• CPC Class Codes: G10L 17/06 Decision making techniques; Pattern matching strategies
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 2015/221 Announcement of recognition results
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command

Abstract

The present disclosure provides a voice recognition method for use in an electronic apparatus comprising a voice input module. The method comprises: receiving voice data by the voice input module; performing a first pattern voice recognition on the received voice data, including identifying whether the voice data comprises a first voice recognition information; performing a second pattern voice recognition on the voice data if the voice data comprises the first voice recognition information; and performing or refusing an operation corresponding to the first voice recognition information according to a result of the second pattern voice recognition. The present disclosure also provides a voice controlling method, an information processing method, and an electronic apparatus.

Claims

1. A voice recognition method for use in an electronic apparatus comprising a voice input module, the method comprising:

receiving voice data by the voice input module;

performing a first pattern voice recognition on the received voice data, including identifying whether the voice data comprises a first voice recognition information;

performing a second pattern voice recognition on the voice data if the voice data comprises the first voice recognition information; and

performing or refusing an operation corresponding to the first voice recognition information according to a result of the second pattern voice recognition.

2. The method according to claim 1, wherein:

the second pattern voice recognition comprises preset-command-related voiceprint authentication; and

the performing or refusing the operation corresponding to the first voice recognition information according to the result of the second pattern voice recognition comprises;

performing the operation corresponding to the first voice recognition information if a result of the preset-command-related voiceprint authentication is SUCCESS; and

refusing the operation corresponding to the first voice recognition information if the result of the preset-command-related voiceprint authentication is FAILURE.

3. The method according to claim 2, wherein:

in case where a second voice recognition information other than the first voice recognition information is identified in the voice data, the method further comprises, prior to performing the operation corresponding to the first voice recognition information;

performing a third pattern voice recognition on the second voice recognition information; and

performing or refusing the operation corresponding to the first voice recognition information and an operation corresponding to the second voice recognition information according to a result of the third pattern voice recognition.

4. The method according to claim 3, wherein:

the third pattern voice recognition comprises preset-command-unrelated voiceprint authentication; and

the performing or refusing the operation corresponding to the first voice recognition information and the operation corresponding to the second voice recognition information according to the result of the third pattern voice recognition comprises;

performing the operation corresponding to the first voice recognition information and the operation corresponding to the second voice recognition information if a result of the preset-command-unrelated voiceprint authentication is SUCCESS; and

refusing the operation corresponding to the first voice recognition information and the operation corresponding to the second voice recognition information if the result of the preset-command-unrelated voiceprint authentication is FAILURE.

5. The method according to claim 4, further comprising:

training a model to be used in the preset-command-unrelated voiceprint authentication using voice data of a specific scene to optimize the model.

6. The method according to claim 6, further comprising:

setting or modifying, according to a first instruction, a set of the first voice recognition information or a correspondence between the first voice recognition information and the operation corresponding to the first voice recognition information.

7. A voice controlling method for use in an electronic apparatus comprising a voice input module, the method comprising:

receiving a voice signal by the voice input module;

matching the voice signal with one or more preset instructions, each of the one or more preset instructions comprising corresponding user-defined voice data; and

performing, if the voice signal matches one of the one or more preset instructions, an operation corresponding to the matched preset instruction.

8. The method according to claim 7, further comprising:

inputting the user-defined voice data;

verifying the input user-defined voice data and storing the verified user-defined voice data in a first storage module as the preset instruction corresponding to a first operation;

transmitting the preset instruction corresponding to the first operation to a predetermined apparatus after the preset instruction corresponding to the first operation is received by the first storage module;

checking whether a model that corresponds to the preset instruction corresponding to the first operation is stored in the predetermined apparatus;

retrieving and storing the model if the model is stored in the predetermined apparatus; and

checking whether the model is stored in the electronic apparatus if the model is not stored in the predetermined apparatus and establishing and storing the model if the model is not stored in the electronic apparatus.

9. The method according to claim 7, further comprising, prior to performing the operation corresponding to the matched preset instruction:

performing a first pattern voice recognition on the voice signal; and

performing or refusing the operation corresponding to the matched preset instruction according to a result of the first pattern voice recognition.

10. The method according to claim 9, wherein:

the first pattern voice recognition comprises performing preset-command-related voiceprint authentication on the voice signal; and

the performing or refusing the operation corresponding to the matched preset instruction according to the result of the first pattern voice recognition comprises;

performing the operation corresponding to the matched preset instruction if a result of the preset-command-related voiceprint authentication is SUCCESS; and

refusing the operation corresponding to the matched preset instruction if the result of the preset-command-related voiceprint authentication is FAILURE.

11. The method according to claim 10, further comprising, prior to performing the operation corresponding to the matched preset instruction:

performing a second pattern voice recognition on a voice recognition information other than the matched preset instruction; and

performing or refusing the operation corresponding to the matched preset instruction and an operation corresponding to the voice recognition information according to a result of the second pattern voice recognition.

12. The method according to claim 11, wherein:

the second pattern voice recognition comprises preset-command-unrelated voiceprint authentication; and

the performing or refusing the operation corresponding to the matched preset instruction and the operation corresponding to the voice recognition information according to the result of the second pattern voice recognition comprises;

performing the operation corresponding to the matched preset instruction and the operation corresponding to the voice recognition information if a result of the preset-command-unrelated voiceprint authentication is SUCCESS; and

refusing the operation corresponding to the matched preset instruction and the operation corresponding to the voice recognition information if the result of the preset-command-unrelated voiceprint authentication is FAILURE.

13. The method according to claim 7, wherein:

each of the one or more preset instructions further comprises corresponding user-defined text data; and

matching the voice signal with the one or more preset instructions comprises;

comparing audio data containing the voice signal with the user-defined voice data in a first preset instruction and determining that a first determination result is MATCH if a similarity degree therebetween exceeds a first threshold;

comparing text data obtained by converting the voice signal with the user-defined text data in the first preset instruction and determining that a second determination result is MATCH if a similarity degree therebetween exceeds a second threshold; and

determining that the voice signal matches the first preset instruction if both the first determination result and the second determination result are MATCH.

14. An information processing method for use in an electronic apparatus, the method comprising:

detecting a voice information;

obtaining a first content information and a first voiceprint information from the voice information;

controlling the electronic apparatus to run a first application corresponding to the first content information; and

controlling the electronic apparatus to run a second application other than the first application if the first voiceprint information is a preset voiceprint information.

15. The method according to claim 14, wherein the obtaining the first content information and the first voiceprint information from the voice information comprises:

obtaining a first reliability degree of the first content information and a second reliability degree of the first voiceprint information, wherein the first reliability degree corresponds to reliability of the first content information and the second reliability degree corresponds to reliability of the first voiceprint information;

selecting a first reliability degree threshold from at least two preset reliability degree thresholds;

determining whether the first reliability degree is larger than the first reliability degree threshold to obtain a first determination result and/or determining whether the second reliability degree is larger than the first reliability degree threshold to obtain a second determination result; and

setting the first content information as a final content information if the first determination result is YES and setting the first voiceprint information as a final voiceprint information if the second determination result is YES,

wherein the first determination result being YES indicates that the first content information is accurate and the second determination result being YES indicates that the first voiceprint information is accurate.

16. The method according to claim 15, wherein the at least two reliability degree thresholds are preset according to at least one of:

contents identifiable by a second obtaining unit of the electronic apparatus and a network condition of the electronic apparatus.

17. The method according to claim 15, further comprising:

after the obtaining the first determination result and/or the second determination result;

transmitting the voice information to a second voice recognition device connected with the electronic apparatus if the first determination result and/or the second determination result is NO, the second voice recognition device being configured for identifying the voice information to obtain a second content information if the first determination result is NO and to obtain a second voiceprint information if the second determination result is NO; and

receiving the obtained second content information and/or second voiceprint information from the second voice recognition device and setting the obtained second content information and/or second voiceprint information as the final content information or the final voiceprint information, respectively;

or

after the detecting the voice information;

transmitting the voice information to a second voice recognition device connected with the electronic apparatus, the second voice recognition device being configured for identifying the voice information to obtain a second content information and/or a second voiceprint information; and

receiving the second content information from the second voice recognition device if the first determination result is NO and receiving the second voiceprint information from the second voice recognition device if the second determination result is NO within a preset time period, if the first determination result and/or the second determination result are NO.

18. The method according to claim 15, further comprising, after the controlling the electronic apparatus to run the first application corresponding to the first content information and controlling the electronic apparatus to run the second application other than the first application if the first voiceprint information is the preset voiceprint information:

obtaining a first authentication information corresponding to the first application by running the second application, the first authentication information being stored in the electronic apparatus and used for authentication of the first application.

19. The method according to claim 18, wherein the obtaining the first authentication information corresponding to the first application by running the second application comprises:

generating a dynamic password through running the second application by the electronic apparatus;

detecting a second voice information produced when a user is reading the dynamic password;

determining whether the second voice information satisfies a preset condition to obtain a third determination result; and

obtaining the first authentication information if the third determination result is YES.

20. The method according to claim 19, wherein the determining whether the second voice information satisfies the preset condition to obtain the third determination result comprises:

obtaining a third content information and a third voiceprint information from the second voice information;

determining whether the third content information is the dynamic password to obtain a fourth determination result;

determining whether the third voice information is the preset voiceprint information if the fourth determination result is YES; and

determining the third determination result to be YES if the fourth determination result is YES and the third voiceprint information is the preset voiceprint information and otherwise determining the third determination result to be NO.