SPEECH INTERACTION METHOD AND APPARATUS

• US 20150206533A1
• Filed: 01/20/2015
• Published: 07/23/2015
• Est. Priority Date: 01/20/2014
• Status: Active Grant

• Current Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Inventors: JIN, Hongbo, JIANG, Zhuolin
• Granted Patent: US 9,583,101 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06V 40/172 Classification, e.g. identification
• CPC Class Codes: G06V 40/70 Multimodal biometrics, e.g. combining information from different biometric modalities
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/25 using position of the lips, movement of the lips or face analysis
• CPC Class Codes: G10L 17/10 Multimodal systems, i.e. based on the integration of multiple recognition engines or fusion of expert systems
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 17/26 Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 2015/227 of the speaker; Human-factor methodology

Abstract

The present invention discloses a speech interaction method and apparatus, and pertains to the field of speech processing technologies. The method includes: acquiring speech data of a user; performing user attribute recognition on the speech data to obtain a first user attribute recognition result; performing content recognition on the speech data to obtain a content recognition result of the speech data; and performing a corresponding operation according to at least the first user attribute recognition result and the content recognition result, so as to respond to the speech data. According to the present invention, after speech data is acquired, user attribute recognition and content recognition are separately performed on the speech data to obtain a first user attribute recognition result and a content recognition result, and a corresponding operation is performed according to at least the first user attribute recognition result and the content recognition result.

Claims

1. A speech interaction method, comprising:

acquiring speech data of a user;

performing user attribute recognition on the speech data to obtain a first user attribute recognition result;

performing content recognition on the speech data to obtain a content recognition result of the speech data; and

performing a corresponding operation according to at least the first user attribute recognition result and the content recognition result, so as to respond to the speech data.

2. The method according to claim 1, further comprising:

collecting a user image; and

when it is detected that the number of people in the user image is a preset value, performing the step of performing user attribute recognition on the speech data.

3. The method according to claim 2, after it is detected that the number of people in the user image is a preset value, the method further comprising:

extracting face data from the user image; and

performing face recognition on the face data to obtain a second user attribute recognition result.

4. The method according to claim 3, wherein the performing a corresponding operation according to at least the first user attribute recognition result and the content recognition result comprises:

weighting the first user attribute recognition result and the second user attribute recognition result to obtain a final user attribute recognition result; and

performing a corresponding operation according to the final user attribute recognition result and the content recognition result.

5. The method according to claim 4, wherein the performing a corresponding operation according to the final user attribute recognition result and the content recognition result comprises:

determining vocabulary content corresponding to the final user attribute recognition result; and

performing a corresponding operation according to the content recognition result and the vocabulary content corresponding to the final user attribute recognition result.

6. The method according to claim 5, wherein before the determining vocabulary content corresponding to the final user attribute recognition result, the method further comprises:

presetting a user attribute, wherein the user attribute comprises at least a gender attribute and an age attribute; and

presetting multiple vocabularies corresponding to the gender attribute and multiple vocabularies corresponding to the age attribute, and storing correspondences; and

the determining vocabulary content corresponding to the final user attribute recognition result comprises;

searching, in a preset correspondence between the gender attribute and a vocabulary and a preset correspondence between the age attribute and a vocabulary, for a vocabulary corresponding to the final user attribute recognition result, and using a found vocabulary as the vocabulary content corresponding to the final user attribute recognition result.

7. The method according to claim 1, wherein before the performing user attribute recognition on the speech data to obtain a first user attribute recognition result of the speech data, the method further comprises:

presetting a correspondence between a spectrum signature and a user attribute recognition result, wherein the user attribute recognition result comprises at least a gender attribute recognition result and an age attribute recognition result; and

the performing user attribute recognition on the speech data to obtain a first user attribute recognition result of the speech data comprises;

performing frequency domain transformation on the speech data to obtain a spectrum signature of the speech data; and

searching, in a preset correspondence between each spectrum signature and each user attribute recognition result, for a user attribute recognition result corresponding to the spectrum signature of the speech data, and using a found user attribute recognition result as the first user attribute recognition result of the speech data.

8. A speech interaction apparatus, comprising:

an acquiring module, configured to acquire speech data of a user;

a user attribute recognition module, configured to perform user attribute recognition on the speech data to obtain a first user attribute recognition result;

a content recognition module, configured to perform content recognition on the speech data to obtain a content recognition result of the speech data; and

an executing module, configured to perform a corresponding operation according to at least the first user attribute recognition result and the content recognition result, so as to respond to the speech data.

9. The apparatus according to claim 8, wherein the apparatus further comprises:

a collecting module, configured to collect a user image, wherein

the user attribute recognition module is configured to, when it is detected that the number of people in the user image is a preset value, perform the step of performing user attribute recognition on the speech data.

10. The apparatus according to claim 9, wherein the apparatus further comprises:

an extracting module, configured to extract face data from the user image; and

a face recognition module, configured to perform face recognition on the face data to obtain a second user attribute recognition result.

11. The apparatus according to claim 10, wherein the executing module comprises:

a weighting unit, configured to weight the first user attribute recognition result and the second user attribute recognition result to obtain a final user attribute recognition result; and

an executing unit, configured to perform a corresponding operation according to the final user attribute recognition result and the content recognition result.

12. The apparatus according to claim 11, wherein the executing unit comprises:

a determining subunit, configured to determine vocabulary content corresponding to the final user attribute recognition result; and

an executing subunit, configured to perform a corresponding operation according to the content recognition result and the vocabulary content corresponding to the final user attribute recognition result.

13. The apparatus according to claim 12, wherein the apparatus further comprises:

a user attribute setting module, configured to preset a user attribute, wherein the user attribute comprises at least a gender attribute and an age attribute; and

a vocabulary correspondence setting module, configured to preset multiple vocabularies corresponding to the gender attribute and multiple vocabularies corresponding to the age attribute, and store correspondences;

wherein

the determining subunit is configured to search, in a preset correspondence between the gender attribute and a vocabulary and a preset correspondence between the age attribute and a vocabulary, for a vocabulary corresponding to the final user recognition result, and use a found vocabulary as the vocabulary content corresponding to the final user attribute recognition result.

14. The apparatus according to claim 8, wherein the apparatus further comprises:

a spectrum signature setting module, configured to preset a correspondence between a spectrum signature and a user attribute recognition result, wherein the user attribute recognition result comprises at least a gender attribute recognition result and an age attribute recognition result, wherein

the user attribute recognition module is configured to perform frequency domain transformation processing on the speech data to obtain a spectrum signature of the speech data; and

search, in a preset correspondence between each spectrum signature and each user attribute recognition result, for a user attribute recognition result corresponding to the spectrum signature of the speech data, and use a found user attribute recognition result as the first user attribute recognition result of the speech data.