METHOD AND APPARATUS FOR SPEECH RECOGNITION

• US 20160210965A1
• Filed: 01/12/2016
• Published: 07/21/2016
• Est. Priority Date: 01/19/2015
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Mun, Minyoung, Choi, YoungSang
• Granted Patent: US 9,953,647 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/32 Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 21/0216 characterised by the method used for estimating noise

Abstract

A method and apparatus for speech recognition are provided. The method and the apparatus calculate signal to noise ratios (SNRs) of speech signals from a user received at speech recognition apparatuses. The method and the apparatus recognize a reference speech signal having a maximum SNR among the SNRs.

Claims

1. A speech recognition method, comprising:

calculating signal to noise ratios (SNRs) of speech signals from a user received at speech recognition apparatuses; and

recognizing a reference speech signal having a maximum SNR among the SNRs.

2. The method of claim 1, wherein the calculating of the SNRs comprises calculating the SNRs by separating noise and a speech of the user comprised in the speech signals.

3. The method of claim 1, wherein the recognizing of the reference speech signal comprises:

determining a speech recognition apparatus to recognize the reference speech signal among the speech recognition apparatuses based on an amount of available resources of a reference speech recognition apparatus receiving the reference speech signal; and

recognizing the reference speech signal in the speech recognition apparatus.

4. The method of claim 3, wherein the recognizing of the reference speech signal comprises recognizing the reference speech signal in the reference speech recognition apparatus in response to the amount of available resources being greater than a threshold value.

5. The method of claim 3, wherein the recognizing of the reference speech signal comprises:

transmitting a feature value extracted from the reference speech signal or the reference speech signal to another speech recognition apparatus in response to the amount of available resources being less than the threshold value; and

recognizing the speech signal in the another speech recognition apparatus.

6. The method of claim 1, wherein the calculating of the SNRs comprises, when an speech recognition apparatus of which a battery condition is less than a threshold value is present among the speech recognition apparatuses, calculating an SNR by transmitting a speech signal received at the speech recognition apparatus of which the battery condition is less than the threshold value to another speech recognition apparatus.

7. The method of claim 1, wherein the speech recognition apparatus having the maximum SNR is determined by another speech recognition apparatus having received the SNRs from the speech recognition apparatuses or a communication among the speech recognition apparatuses.

8. The method of claim 1, further comprising:

determining whether the user issues a command to any speech recognition apparatus among the speech recognition apparatuses; and

receiving the speech signals through the speech recognition apparatuses in response to a determination that the user issues the command to any speech recognition apparatus among the speech recognition apparatuses.

9. The method of claim 1, further comprising:

verifying a user command by recognizing the speech signals and transmitting the user command to an speech recognition apparatus corresponding to the command,

wherein the speech recognition apparatus corresponding to the command executes the command.

10. The method of claim 1, wherein the speech recognition apparatuses are positioned at different positions adjacent to the user.

11. The method of claim 1, wherein the speech recognition apparatuses correspond to objects in an Internet of Things (IoT) environment.

12. The method of claim 1, wherein the speech recognition apparatuses comprises at least one of a smartphone, a smart television, a smart watch, a smart washing machine, a smart camera, a smart audio, a smart bicycle, smart glasses, a remote controller, a smart refrigerator, a smart air conditioner, a laptop, a personal digital assistant (PDA), and a tablet.

13. A non-transitory computer-readable storage medium comprising a program comprising instructions to cause a computer to perform the method of claim 1.

14. A speech recognition apparatus, comprising:

a microphone configured to receive a reference speech signal from a user; and

a processor configured to recognize the reference speech signal in response to a signal to noise ratio (SNR) calculated from the speech signal being greater than an SNR of the speech signal received at another speech recognition apparatus.

15. The apparatus of claim 14, wherein the processor is configured to calculate the SNR by separating a noise and a speech of the user comprised in the reference speech signal.

16. The apparatus of claim 14, wherein the processor is configured to determine whether the reference speech signal is to be recognized based on an amount of available resources or by transmitting the reference speech signal to the another speech recognition apparatus.

17. The apparatus of claim 16, wherein the processor is configured to recognize the reference speech signal in response to the amount of available resources being greater than a threshold value.

18. The apparatus of claim 16, wherein the processor is configured to transmit a feature value extracted from the reference speech signal or the reference speech signal to the another speech recognition apparatus in response to the amount of available resources being less than a threshold value, and the another speech recognition apparatus is configured to recognize the reference speech signal.

19. The apparatus of claim 14, wherein the processor is configured to determine whether the user issues a command to the speech recognition apparatus or to the another speech recognition apparatus, and the microphone is configured to receive the reference speech signal in response to a determination that the user issues the command to the speech recognition apparatus or to the another speech recognition apparatus.

20. The apparatus of claim 14, wherein the speech recognition apparatus is configured to correspond to an object in an Internet of Things (IoT) environment.

21. The apparatus of claim 14, wherein, in response to the SNR being a high reference speech signal, the processor performs speech recognition at a high recognition rate.

22. The apparatus of claim 14, wherein the another speech recognition apparatus comprises a battery condition better than the speech recognition apparatus or disposed closest to the speech recognition apparatus.