SPEECH RECOGNITION SYSTEM AND SPEECH RECOGNITION METHOD THEREOF

• US 20160267913A1
• Filed: 03/11/2016
• Published: 09/15/2016
• Est. Priority Date: 03/13/2015
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: KIM, Tae-yoon, KIM, Sang-ha, HAN, Chang-woo, LEE, Jae-won
• Granted Patent: US 10,699,718 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 1/3215 Monitoring of peripheral devices
• CPC Class Codes: G06F 1/3231 Monitoring the presence, absence or movement of users
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 17/20 Pattern transformations or operations aimed at increasing system robustness, e.g. against channel noise or different working conditions
• CPC Class Codes: G10L 17/24 the user being prompted to utter a password or a predefined phrase
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A device detects a wake-up keyword from a received speech signal of a user by using a wake-up keyword model, and transmits a wake-up keyword detection/non-detection signal and the received speech signal of the user to a speech recognition server. The speech recognition server performs a recognition process on the speech signal of the user by setting a speech recognition model according to the detection or non-detection of the wake-up keyword.

Claims

1. A device comprising:

an audio input receiver configured to receive a speech signal of a user;

a memory configured to store a wake-up keyword model;

a communicator configured to communicate with a speech recognition server; and

a processor configured to detect a wake-up keyword from the speech signal of the user by using the wake-up keyword model in response to receiving the speech signal of the user, transmit a wake-up keyword detection/non-detection signal and the speech signal of the user to the speech recognition server via the communicator, receive, from the speech recognition server via the communicator, a speech recognition result of the speech signal based on the wake-up keyword detection/non-detection signal, and control the device according to the speech recognition result.

2. The device of claim 1, further comprising an environment information sensor configured to sense environment information about the device,

wherein the wake-up keyword model is one of a plurality of wake-up keyword models which are based on various environment information, and

the processor is further configured to detect the wake-up keyword from the received speech signal of the user by using the wake-up keyword model corresponding to the sensed environment information of the device, among the plurality of wake-up keyword models.

3. The device of claim 2, wherein the environment information includes position information of the device.

4. The device of claim 2, wherein the processor is further configured to acquire identification information of the user based on the received speech signal of the user, transmit the identification information of the user and the environment information of the device to the speech recognition server via the communicator, and receive, from the speech recognition server, the speech recognition result based on at least two of the wake-up keyword detection/non-detection signal, the identification information of the user, and the environment information of the device.

5. The device of claim 1, wherein when an audio signal for registering the wake-up keyword model is received through the audio input unit, the processor is further configured to determine whether the audio signal is valid as the wake-up keyword model, based on a matching rate of a recognition of the audio signal.

6. The device of claim 1, further comprising a user input receiver configured to receive a user input,

wherein the memory is configured to store candidate wake-up keyword models, and

the processor is further configured to register a candidate wake-up keyword model, which is selected among the stored candidate wake-up keyword models, according to the user input received through the user input receiver.

7. The device of claim 1, wherein the processor is further configured to receive the wake-up keyword model from the speech recognition server via the communicator and register the received wake-up keyword model in the memory.

8. The device of claim 1, wherein the processor is further configured to generate the wake-up keyword model based on the speech signal of the user, which is received through the audio input receiver, and register the generated wake-up keyword model in the memory.

9. A speech recognition server comprising:

a communicator configured to communicate with a device;

a memory configured to store a wake-up keyword model and a speech recognition model; and

a processor configured to set a speech recognition model combined with the wake-up keyword model in response to receiving a wake-up keyword detection/non-detection signal and a speech signal of a user from the device via the communicator, recognize the speech signal of the user by using the set speech recognition model and generate a speech recognition result of the speech signal, remove the wake-up keyword from the speech recognition result of the speech signal of the user, and transmit the speech recognition result, from which the wake-up keyword has been removed, to the device via the communicator.

10. The speech recognition server of claim 9, wherein the wake-up keyword model is one of a plurality of wake-up keyword models which are based on various environment information, and

the processor is further configured to set the speech recognition model combined with the wake-up keyword model corresponding to environment information of the device, among the plurality of wake-up keyword models, in response to receiving the environment information of the device via the communicator, and generate the speech recognition result based on the environment information.

11. The speech recognition server of claim 10, wherein the processor is further configured to set the speech recognition model combined with identification information of the user and the wake-up keyword model corresponding to the environment information of the device, in response to receiving the identification information of the user via the communicator, and generate the speech recognition result further based on the identification information.

12. A speech recognition method, which is performed by a device, the speech recognition method comprising:

receiving a speech signal of a user;

detecting a wake-up keyword from the received speech signal of the user by using a wake-up keyword model;

transmitting a wake-up keyword detection/non-detection signal and the speech signal of the user to a speech recognition server;

receiving a result of recognizing the speech signal of the user from the speech recognition server, based on the wake-up keyword detection/non-detection signal; and

controlling the device according to the result of recognizing the speech signal of the user.

13. The speech recognition method of claim 12, wherein the wake-up keyword model is one of a plurality of wake-up keyword models which are based on various environment information,

the speech recognition method further comprises detecting environment information about the device, and transmitting the detected environment information of the device to the speech recognition server, and

the detecting the wake-up keyword comprises detecting the wake-up keyword from the speech signal of the user by using the wake-up keyword model corresponding to the detected environment information of the device, among the plurality of wake-up keyword models.

14. The speech recognition method of claim 13, further comprising acquiring identification information of the user based on the speech signal of the user,

wherein the transmitting comprises transmitting the identification information of the user and the environment information of the device to the speech recognition server, and

the receiving the result of recognizing the speech signal comprises receiving, from the speech recognition server, the result of recognizing the speech signal based on at least two of the wake-up keyword detection/non-detection signal, the identification information of the user, and the environment information of the device.

15. A non-transitory computer-readable recording medium having stored thereon a program which, when executed by a computer system, causes the computer system to perform the method of claim 12.