The 4K audio signal noise presents a complex state and may lead to an unstable state under the impact response.I-dentifying and labeling the signal noise is a challenge.This paper proposes a method of non-stationary noise filtering for 4K high-definition all-media broadcasting vehicle audio signals.The amplitude frequency response feature extraction method is used to calculate the position scale parameters of non-stationary noise signals,split audio signal segments,and construct an impact response function.With the support of the impact response function,particle filtering is used to extract non-stationary noise features.Based on the feature extraction results,the audio signals are segmented in the time domain.Wigner transform is used to analyze the time-frequency domain of analytic signal of audio signal,and the marking function of non-stationary noise is con-structed to complete the detection of non-stationary noise.After encoding the non-stationary noise,the corresponding analytic signal is obtained,and the non-stationary noise filtering is completed.The experimental results show that the proposed method can improve the signal-to-noise ratio of audio signals and shorten the filtering time while improving the filtering effect of non-stationary noise,which indicates that the non-stationary noise filtering method studied has higher practical application signifi-cance.
维普
万方数据