Warning a user when voice input to a device is likely to fail because of background or other noise

• US 8,438,023 B1
• Filed: 08/17/2012
• Issued: 05/07/2013
• Est. Priority Date: 09/30/2011
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Hamilton, Robert William, Bringert, Bjorn Erik, LeBeau, Michael J., Byrne, William J., Jitkoff, John Nicholas
• Primary Examiner(s): PULLIAS, JESSE SCOTT
• Application Number:
  US13/588,054
• Time in Patent Office: 263 Days
• Field of Search: 704/231, 704/233, 704/240, 704/274
• US Class Current: 704/233
• CPC Class Codes: G10L 15/01 Assessment or evaluation of speech recognition systems
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command

Abstract

Methods and devices are disclosed for notifying a user of a likelihood of successful recognition in an environment by a voice recognition application. In one embodiment, the method includes a device recording a noise sample in an environment and making a comparison of the noise sample and at least one predetermined threshold. The method further includes, based on the comparison, determining a likelihood of successful recognition in the environment by a voice recognition application, and triggering a notification indicating the likelihood. In another embodiment, the device includes a microphone configured to record a noise sample in an environment, a processor, and data storage comprising instructions executable by the processor to make a comparison of the noise sample and at least one predetermined threshold, based on the comparison, determine a likelihood of successful recognition by a voice recognition application, and trigger a notification indicating the likelihood.

Claims

1. A method comprising:

a device recording a noise sample in an environment;

making a comparison of the noise sample and at least one predetermined threshold;

based on the comparison, determining a likelihood of successful recognition in the environment by a voice recognition application comprising determining whether the likelihood of successful recognition is high, medium, or low; and

triggering a notification indicating the likelihood, comprising;

after determining that the likelihood of successful recognition is high, triggering a notification that indicates that the likelihood of successful recognition is high,

after determining that the likelihood of successful recognition is medium, triggering a notification that indicates that the likelihood of successful recognition is medium, and

after determining that the likelihood of successful recognition is low, triggering a notification that indicates that the likelihood of successful recognition is low.

2. The method of claim 1, wherein the device recording the noise sample comprises the device recording the noise sample in response to a trigger.

3. The method of claim 2, wherein the trigger comprises at least one of (i) receiving a request from a user for the voice recognition application, (ii) detecting that the device is located in the environment, (iii) detecting an event on the device, and (iv) detecting that the device is connected to a power source.

4. The method of claim 1, wherein the device recording the noise sample comprises the device continuously recording a plurality of noise samples including the noise sample.

5. The method of claim 1, wherein:

the noise sample has a volume;

the at least one predetermined threshold includes a predetermined volume threshold; and

making the comparison comprises making a comparison of the volume and the predetermined volume threshold.

6. The method of claim 1, wherein the at least one predetermined threshold is predetermined based on at least one predetermined noise sample.

7. The method of claim 6, wherein:

the noise sample has a volume and a noise type;

the at least one predetermined threshold includes a predetermined volume threshold;

the at least one predetermined noise sample has a predetermined noise type; and

making the comparison comprises making a comparison of at least one of (i) the volume and the predetermined volume threshold and (ii) the noise type and the predetermined noise type.

8. The method of claim 1, further comprising:

determining a type of the environment; and

based on the type of the environment, selecting the at least one predetermined noise sample from a plurality of predetermined noise samples.

9. The method of claim 1, wherein the notification comprises at least one of a visual notification, an audible notification, and a tactile notification.

10. A device comprising:

at least one microphone configured to record a noise sample in an environment;

at least one processor; and

data storage comprising instructions executable by the processor to;

make a comparison of the noise sample and at least one predetermined threshold;

based on the comparison, determine a likelihood of successful recognition in the environment by a voice recognition application comprising determining whether the likelihood of successful recognition is high, medium, or low; and

trigger a notification indicating the likelihood, comprising;

after determining that the likelihood of successful recognition is high, triggering a notification that indicates that the likelihood of successful recognition is high,

after determining that the likelihood of successful recognition is medium, triggering a notification that indicates that the likelihood of successful recognition is medium, and

after determining that the likelihood of successful recognition is low, triggering a notification that indicates that the likelihood of successful recognition is low.

11. The device of claim 10, further comprising at least one of a display and a light configured to display the notification.

12. The device of claim 10, further comprising a speaker configured to emit the notification.

13. The device of claim 10, further comprising a vibration motor configured to emit the notification.

14. The device of claim 10, wherein the device is selected from the group consisting of a mobile telephone, a smartphone, a tablet computer, a laptop computer, a wearable computer, a desktop computer, a computer embedded in a vehicle, and a computer embedded in an appliance.

15. The device of claim 10, wherein the at least one microphone is further configured to record a voice input for use in the voice recognition application.

16. A device comprising:

at least one microphone configured to record a noise sample in an environment;

an indicator;

at least one processor; and

data storage comprising instructions executable by the processor to;

make a comparison of the noise sample and at least one predetermined threshold;

based on the comparison, determine a likelihood of successful recognition in the environment by a voice recognition application comprising determining whether the likelihood of successful recognition is high, medium, or low; and

trigger, via the indicator, a textual notification indicating the likelihood and a graphical notification indicating the likelihood, comprising;

after determining that the likelihood of successful recognition is high, triggering a textual notification that indicates that the likelihood of successful recognition is high and a graphical notification that indicates that the likelihood of successful recognition is high,

after determining that the likelihood of successful recognition is medium, triggering a textual notification that indicates that the likelihood of successful recognition is medium and a graphical notification that indicates that the likelihood of successful recognition is medium, and

after determining that the likelihood of successful recognition is low, triggering a textual notification that indicates that the likelihood of successful recognition is low and a graphical notification that indicates that the likelihood of successful recognition is low.

17. The device of claim 16, wherein the textual notification comprises at least one of a qualitative indication of the likelihood and a quantitative indication of the likelihood.

18. The device of claim 16, wherein the graphical notification comprises at least one of a qualitative indication of the likelihood and a quantitative indication of the likelihood.

19. A non-transitory computer readable medium having stored therein instructions executable by a computing device to cause the computing device to perform functions comprising:

recording a noise sample in an environment;

making a comparison of the noise sample and at least one predetermined threshold;

based on the comparison, determining a likelihood of successful recognition in the environment by a voice recognition application comprising determining whether the likelihood of successful recognition is high, medium, or low; and

triggering a notification indicating the likelihood, comprising;

after determining that the likelihood of successful recognition is high, triggering a notification that indicates that the likelihood of successful recognition is high,

after determining that the likelihood of successful recognition is medium, triggering a notification that indicates that the likelihood of successful recognition is medium, and

after determining that the likelihood of successful recognition is low, triggering a notification that indicates that the likelihood of successful recognition is low.

20. The non-transitory computer readable medium of claim 19, wherein recording the noise sample comprises recording the noise sample in response to a trigger.

21. The non-transitory computer readable medium of claim 19, wherein recording the noise sample comprises continuously recording a plurality of noise samples including the noise sample.

22. The non-transitory computer readable medium of claim 19, wherein the at least one predetermined threshold is predetermined based on at least one predetermined noise sample.

23. The non-transitory computer readable medium of claim 19, wherein the notification comprises at least one of a visual notification, an audible notification, and a tactile notification.