Artificial intelligence device for diagnosing failure and method thereof

Patent metadata

Field	Value
Publication number	US-11330384-B2
Application number	US-201916707102-A
Country	US
Kind code	B2
Filing date	Dec 9, 2019
Priority date	Nov 8, 2019
Publication date	May 10, 2022
Grant date	May 10, 2022

Abstract

Disclosed herein is an artificial intelligence device including a communication interface configured to receive a reference signal in a time domain for echo cancellation of an external device and a processor configured to determine whether a spectral density at at least one test target frequency is less than or equal to a preset spectral density reference value at at least one test target time point and determine that a failure occurs in an audio component when the spectral density is less than or equal to the preset spectral density reference value.

First claim

What is claimed is:

1. An artificial intelligence device comprising:
a communication interface configured to receive a reference signal in a time domain for echo cancellation of an external device; and
a processor configured to determine whether a spectral density at at least one test target frequency is less than or equal to a preset spectral density reference value at at least one test target time point and determine that a failure has occurred in an audio component when the spectral density is less than or equal to the preset spectral density reference value,
wherein the communication interface receives a plurality of audio recording signals obtained by recording an audio signal output from a speaker of the external device by a plurality of microphones of the external device, and
wherein the processor is configured to perform comparison with a spectral density difference in a frequency band in a predetermined range with respect to the plurality of audio recording signals, and determine that a failure has occurred in at least one of the plurality of microphones of the external device when the spectral density difference is greater than or equal to a preset reference value.

2. The artificial intelligence device of claim 1 , wherein the processor is configured to:
convert the reference signal in the time domain into a reference signal in a frequency domain, obtain a start frequency at which the reference signal in the time domain starts, and determine a time point at which a spectral density of the start frequency starts to exceed the preset spectral density reference value with respect to the reference signal in the frequency domain as a recording start time point at which the reference signal in the time domain starts to be recorded, and
determine whether the spectral density at the at least one test target frequency is less than or equal to the preset spectral density reference value at a test target time point corrected based on the recording start time point and determine that a failure has occurred in the audio component of the external device when the spectral density is less than or equal to the preset spectral density reference value.

3. The artificial intelligence device of claim 2 , wherein the processor is configured to determine that a failure has occurred in an audio cable of the external device when the spectral density is less than or equal to the preset spectral density reference value, and
wherein the communication interface transmits a failure occurrence notification for the audio cable to the external device.

4. The artificial intelligence device of claim 1 ,
wherein the processor is configured to detect an abnormal signal which deviates from a preset maximum inclination value with respect to an audio recording signal of the plurality of audio recording signals, determine whether there is an abnormality in the audio recording signal according to whether a predetermined number or more of abnormal signals are detected, and determine that a failure has occurred in the audio component of the external device when it is determined that there is the abnormality in the audio recording signal.

5. The artificial intelligence device of claim 4 , wherein the processor is configured to measure a maximum amplitude of the audio recording signal to set the maximum amplitude to the maximum inclination value, and detect a signal of which a differential value of the audio recording signal exceeds the maximum inclination value as the abnormal signal.

6. The artificial intelligence device of claim 4 , wherein the processor is configured to determine that a failure has occurred in an audio board of the external device when it is determined that there is the abnormality in the audio recording signal, and
wherein the communication interface transmits a failure occurrence notification for the audio board to the external device.

7. The artificial intelligence device of claim 4 , wherein the audio signal is a signal with a constant amplitude level that lasts at a predetermined frequency.

8. The artificial intelligence device of claim 1 , wherein the plurality of audio recording signals are each a sweep signal of which a frequency increases at a predetermined frequency rate per unit time or a white noise signal comprising a frequency component in a predetermined range.

9. A method of diagnosing a failure, the method comprising:
receiving a reference signal in a time domain for echo cancellation of an external device;
determining whether a spectral density at at least one test target frequency is less than or equal to a preset spectral density reference value at at least one test target time point; and
determining that a failure has occurred in an audio component when the spectral density is less than or equal to the preset spectral density reference value,
wherein the method further comprises:
receiving a plurality of audio recording signals obtained by recording an audio signal output from a speaker of the external device by a plurality of microphones of the external device;
performing comparison with a spectral density difference in a frequency band in a predetermined range with respect to the plurality of audio recording signals; and
determining that a failure has occurred in at least one of the plurality of microphones of the external device when the spectral density difference is greater than or equal to a preset reference value.

10. The method of claim 9 , wherein the determining whether the spectral density is less than or equal to the preset spectral density reference value includes
converting the reference signal in the time domain into a reference signal in a frequency domain,
obtaining a start frequency at which the reference signal in the time domain starts,
determining a time point at which a spectral density of the start frequency starts to exceed the preset spectral density reference value with respect to the reference signal in the frequency domain as a recording start time point at which the reference signal in the time domain starts to be recorded,
determining whether the spectral density at the at least one test target frequency is less than or equal to the preset spectral density reference value at a test target time point corrected based on the recording start time point, and
determining that a failure has occurred in the audio component of the external device when the spectral density is less than or equal to the preset spectral density reference value.

11. The method of claim 10 , wherein the determining that the failure has occurred in the audio component of the external device includes determining that a failure has occurred in an audio cable of the external device when the spectral density is less than or equal to the preset spectral density reference value, and
wherein the method further comprises transmitting a failure occurrence notification for the audio cable to the external device.

12. The method of claim 9 , further comprising:
detecting an abnormal signal which deviates from a preset maximum inclination value with respect to an audio recording signal of the plurality of audio recording signals;
determining whether there is an abnormality in the audio recording signal according to whether a predetermined number or more of abnormal signals are detected; and
determining that a failure has occurred in the audio component of the external device when it is determined that there is the abnormality in the audio recording signal.

13. The method of claim 12 , wherein the detecting the abnormal signal includes
measuring a maximum amplitude of the audio recording signal to set the maximum amplitude to the maximum in

Assignees

• Lg Electronics Inc

Inventors

• Lee Junmin
• Lee Inho

Classifications

• H04R3/005

  for combining the signals of two or more microphones (specially adapted for hearing aids H04R25/407) · CPC title

• G06N3/08

  Learning methods · CPC title

• G10L2021/02082

  the noise being echo, reverberation of the speech · CPC title

• G10L25/18

  the extracted parameters being spectral information of each sub-band · CPC title

• H04R29/005 Primary

  Microphone arrays · CPC title