Speech Separation Algorithm Using Gated Recurrent Network Based on Microphone Array

Speech separation is an active research topic that plays an important role in numerous applications,such as speaker recognition,hearing pros-thesis,and autonomous robots.Many algorithms have been put forward to improve separation performance.However,speech separation in reverberant noisy environment is still a challenging task.To address this,a novel speech separation algorithm using gate recurrent unit(GRU)network based on microphone array has been proposed in this paper.The main aim of the proposed algorithm is to improve the separation performance and reduce the computational cost.The proposed algorithm extracts the sub-band steered response power-phase transform(SRP-PHAT)weighted by gammatone filter as the speech separation feature due to its discriminative and robust spatial position in formation.Since the GRU net work has the advantage of processing time series data with faster training speed and fewer training parameters,the GRU model is adopted to process the separation featuresof several sequential frames in the same sub-band to estimate the ideal Ratio Masking(IRM).The proposed algorithm decomposes the mixture signals into time-frequency(TF)units using gammatone filter bank in the frequency domain,and the target speech is reconstructed in the frequency domain by masking the mixture signal according to the estimated IRM.The operations of decomposing the mixture signal and reconstructing the target signal are completed in the frequency domain which can reduce the total computational cost.Experimental results demonstrate that the proposed algorithm realizes omnidirectional speech sep-aration in noisy and reverberant environments,provides good performance in terms of speech quality and intelligibility,and has the generalization capacity to reverberate.

Percentiles Estimation for a Response Distribution

To estimate percentiles of a response distribution, the transformed response rule of Wetherill and Robbins-Monro sequential design were proposed under Log-Logistic model. Based on responses data, a necessary and sufficient condition for the existence of maximum likelihood estimators and then the calculating formula were presented. After a simulation study, the proposed approach was applied to 65# detonator. Numerical results showed that estimators of percentiles from the proposed approach are robust to the parametric models lacking information on the original response distribution.

Sound Source Localization Based on SRP-PHAT Spatial Spectrum and Deep Neural Network

Microphone array-based sound source localization(SSL)is a challenging task in adverse acoustic scenarios.To address this,a novel SSL algorithm based on deep neural network(DNN)using steered response power-phase transform(SRP-PHAT)spatial spectrum as input feature is presented in this paper.Since the SRP-PHAT spatial power spectrum contains spatial location information,it is adopted as the input feature for sound source localization.DNN is exploited to extract the efficient location information from SRP-PHAT spatial power spectrum due to its advantage on extracting high-level features.SRP-PHAT at each steering position within a frame is arranged into a vector,which is treated as DNN input.A DNN model which can map the SRP-PHAT spatial spectrum to the azimuth of sound source is learned from the training signals.The azimuth of sound source is estimated through trained DNN model from the testing signals.Experiment results demonstrate that the proposed algorithm significantly improves localization performance whether the training and testing condition setup are the same or not,and is more robust to noise and reverberation.

An efficient technique for recovering responses of parameterized structural dynamic problems

In this article,an effective technique is developed to efficiently obtain the output responses of parameterized structural dynamic problems.This technique is based on the conception of reduced basis method and the usage of linear interpolation principle.The original problem is projected onto the reduced basis space by linear interpolation projection,and subsequently an associated interpolation matrix is generated.To ensure the largest nonsingularity,the interpolation matrix needs to go through a timenode choosing process,which is developed by applying the angle of vector spaces.As a part of this technique,error estimation is recommended for achieving the computational error bound.To ensure the successful performance of this technique,the offline-online computational procedures are conducted in practical engineering.Two numerical examples demonstrate the accuracy and efficiency of the presented method.

Analysis of transient viscoelastic response of asphalt concrete using frequency domain approach

The analysis of transient linear viscoelastic response of asphalt concrete （AC） is important for engineering applications. The traditional transient response of AC is analyzed in the time domain by performing complicated convolution integral. The frequency domain approach allows one to determine the transient responses by performing simple multi- plication instead of the complicated convolution integral, and it does not require the time derivative of the input excitation, and thus, the approach could greatly reduce the analysis complexity. This study investigated the frequency domain approach in calculating the transient response by utilizing the discrete Fourier transform technique. The accuracy and effectiveness of the frequency domain approach were verified by comparing the analytical and calculated responses for the standard 3-parameter Maxwell model and by comparing the time and frequency domain solutions for AC. The effect of aliasing of the frequency domain approach can effectively reduce by selecting a small sampling interval for the time domain excitation function. A sampling interval is acceptable as long as the amplitude of the Fourier transformed excitation is close to 0 more than half of the sampling rate. The results show that the frequency domain approach provides a simple and accurate way to perform linear viscoelastic analysis of AC.