System for detecting and reducing noise via a microphone array
View Patent ↗A system for detecting noise in a signal received by a microphone array and a method for detecting noise in a signal received by a microphone array is disclosed. The system also provides for the reduction of noise in a signal received by a microphone array and a method for reducing noise in a signal received by a microphone array. The signal to noise ratio in handsfree systems may be improved, particularly in handsfree systems present in a vehicular environment.
1. A computer-implemented method for detecting wind noise, comprising the steps of:
receiving a plurality of signals from a microphone array;
in a first computer process, decomposing the signals into frequency sub-band signals;
in a second computer process, determining time dependent measures for the signals based on the frequency sub-band signals;
in a third computer process, evaluating a predetermined criterion function using the time dependent measures, producing a criterion result that determines if wind noise is present within the signals; and
in a fourth computer process, using results of the evaluating to detect the wind noise according to a predetermined threshold,
wherein the criterion function is a variance of the time dependent measure, and wherein the time dependent measure is determined as a function of the signal powers of the sub-band signals, and the variance is determined from a mean of the signal powers for detecting the wind noise.
2. The method of claim 1 . where the microphone array further comprises at least two microphones.
3. The method of claim 1 where the predetermined criterion function is a predetermined statistical function of the time dependent measures.
4. The method of claim 1 where the decomposing step further comprises digitizing each signal in the plurality of signals and decomposing each digitized signal into a complex- valued frequency sub-band signal using short time discrete Fourier transform.
5. The method of claim 1 where the decomposing step further comprises digitizing each signal in the plurality of signals and decomposing each digitized signal into a complex-valued frequency sub-band signal using a discrete Wavelet transform.
6. The method of claim 1 where the decomposing step further comprises digitizing each signal in the plurality of signals and decomposing each digitized signal into a complex-valued frequency sub-band signal using a filter bank.
7. The method of claim 1 where the decomposing step further comprises sub-sampling each sub-band signal.
8. The method of claim 1 where the evaluating step comprises comparing the criterion function with a predetermined threshold value, where noise will be detected if the criterion function is larger than the predetermined threshold value.
9. The method according to claim 1 , further including taking a logarithm of the time dependent measure.
10. An article, comprising:
a non-transitory computer-readable medium having stored instructions that enable a machine to detect uncorrelated wind noise, comprising instructions to:
receive a plurality of signals from a microphone array;
decompose the signals into frequency sub-band signals;
determine time dependent measures for the signals based on the frequency sub-band signals;
evaluate a predetermined criterion function using the time dependent measures, producing a criterion result that determines if wind noise is present within the signals; and
use results of the evaluating to detect the wind noise according to a predetermined threshold,
wherein the criterion function is a variance of the time dependent measure, and wherein the time dependent measure is determined as a function. of the signal powers of the sub-band signals, and the variance is determined from a mean of the signal powers for detecting the wind noise.