Electronic devices, methods, and computer program products for detecting noise in a signal based on autocorrelation coefficient gradients
View Patent ↗An electronic device can be operated to detect noise, such as wind noise. A microphone signal is generated by a microphone. Autocorrelation coefficients are determined based on the microphone signal. Gradient values are determined from the autocorrelation coefficients. The presence of a noise component in the microphone signal is determined based on the gradient values.
1. A method of operating an electronic device, the method comprising:
generating autocorrelation coefficients from sampled values of a microphone signal that are delayed by a range of delay values;
determining gradient values from the autocorrelation coefficients; and
detecting presence of a noise component in the microphone signal in response to whether any adjacent gradient values transition from positive values to negative values or from negative values to positive values for delay values that are non-zero.
2. The method of claim 1 , wherein detecting the presence of the noise component comprises determining whether any of the gradient values are about zero for delay values that are non-zero.
3. The method of claim 1 , wherein detecting the presence of a noise component comprises detecting presence of wind noise in the microphone signal in response to at least one of the gradient values being equal to zero for delay values that are non-zero.
4. The method of claim 1 , wherein determining the gradient values from the autocorrelation coefficients comprises weighting newer ones of the delayed samples of the microphone signal greater than older ones of the delayed samples of the microphone signal.
5. The method of claim 1 , further comprising applying a noise suppression algorithm to the microphone signal in response to detecting the presence of a noise component in the microphone signal.
6. An electronic device, comprising:
a microphone that is configured to generate a microphone signal;
an autocorrelation unit that is configured to generate autocorrelation coefficients from sampled values of the microphone signal that are delayed by a range of delay values;
a gradient unit that is configured to generate gradient values from the autocorrelation coefficients; and
a noise detector that is configured to detect presence of a noise component in the microphone signal in response to whether any adjacent gradient values transition from positive values to negative values or from negative values to positive values for delay values that are non-zero.
7. The electronic device of claim 6 , wherein the noise detector is configured to detect the presence of a noise component in the microphone signal in response to whether any of the gradient values are about zero for delay values that are non-zero.
8. The electronic device of claim 6 , wherein the noise detector is further configured to apply at least one noise suppression algorithm to the microphone signal to generate a noise suppressed microphone signal in response to detecting the presence of a noise component in the microphone signal.
9. The electronic device of claim 8 , further comprising a transceiver that is configured to transmit the noise suppressed microphone signal.
10. The electronic device of claim 6 , wherein the noise detector is configured to detect the presence of wind noise in the microphone signal in response to at least one of the gradient values being equal to zero for a delay value that is non-zero.
11. The electronic device of claim 6 , wherein the autocorrelation unit is configured to generate autocorrelation coefficients by weighting newer ones of the delayed samples of the microphone signal greater than older ones of the delayed samples of the microphone signal.
12. A computer program product configured to process a microphone signal produced by a microphone in an electronic device, comprising:
a computer readable storage medium having computer readable program code embodied therein, the computer readable program code comprising:
computer readable program code that generates autocorrelation coefficients from sampled values of a microphone signal that are delayed by a range of delay values;
computer readable program code that determines gradient values from the autocorrelation coefficients; and
computer readable program code that detects presence of a noise component in the microphone signal in response to whether any adjacent gradient values transition from positive values to negative values or from negative values to positive values for delay values that are non-zero.
13. The computer program product of claim 12 , wherein the computer readable program code that detects the presence of a noise component comprises computer readable program code that detects the presence of the noise component in the microphone signal in response to whether any of the gradient values are about zero for delay values that are non-zero.
14. The computer program product of claim 12 , wherein the computer readable program code that detects the presence of a noise component comprises computer readable program code that detects presence of wind noise in the microphone signal in response to at least one of the gradient values being equal to zero for delay values that are non-zero.
15. The computer program product of claim 12 , wherein the computer readable program code that determines gradient values comprises computer readable program code that determines the gradient values from the autocorrelation coefficients by weighting newer ones of the delayed samples of the microphone signal greater than older ones of the delayed samples of the microphone signal.