The Experts below are selected from a list of 234 Experts worldwide ranked by ideXlab platform

Anthony Zaknich - One of the best experts on this subject based on the ideXlab platform.

  • Single channel speech enhancement using a 9 Dimensional Noise Estimation algorithm and Controlled Forward March Averaging
    2008 9th International Conference on Signal Processing, 2008
    Co-Authors: Dariush Farrokhi, Roberto Togneri, Anthony Zaknich
    Abstract:

    A post processing technique is proposed to enhance speech in a single channel system. A new noise estimation algorithm is proposed in conjunction with the Controlled Forward March Averaging (CFMA) technique to enhance speech in a single channel non-stationary noisy system. We introduce a 9-Dimensional Noise Estimation (NDNE) algorithm to the Single Channel Speech Estimation (SCSE) system, that updates the estimated noise in 9 frequency sub-bands, by averaging the noisy speech power spectrum using a time and frequency dependent smoothing Factor. A signal presence Probability Factor is calculated by computing the ratio of the noisy speech power spectrum to its local minimum, which is computed by averaging past values of the noisy speech power spectra with a look-ahead Factor. The NDNE uses a non-linear thresholding map as oppose to the conventional linear thresholding. This new algorithm produced an average 7% improvement in 0 and -2.5 dB global SNR in speech corrupted with modified Babble noise. Subjective tests confirmed these results.

  • Speech enhancement of non-stationary noise based on controlled forward moving average
    2007 International Symposium on Communications and Information Technologies, 2007
    Co-Authors: Dariush Farrokhi, Roberto Togneri, Anthony Zaknich
    Abstract:

    A pre and post processing technique is proposed to enhance the speech signal of highly non-stationary noisy speech. The purpose of this research has been to build on current speech enhancement algorithms to produce an improved algorithm for enhancement of speech contaminated with non-stationary babble type noise. The pre processing involves two stages. In stage one, the variance of the noisy speech spectrum is reduced by utilizing the Discrete or Prolate Spheroidal Sequence (DPSS) multi-taper algorithm plus a Controlled Forward Moving Average (CFMA) technique. We introduced the CFMA algorithm to smooth and reduce variance of the estimated non-stationary noise spectrum. In the second stage the noisy speech power spectrum is de-noised by applying Stein's Unbiased Risk Estimator (SURE) wavelet thresholding technique. In the third layer, use is made of a noise estimation algorithm with rapid adaptation for a highly non-stationary noise environment. The noise estimate is updated in three frequency sub-bands, by averaging the noisy speech power spectrum using a frequency dependent smoothing Factor, which is adjusted, based on a signal presence Probability Factor. In the fourth layer a spectral subtraction algorithm is used to enhance the speech signal, by subtracting each estimated noise from the original noisy speech. The new proposed post processing is then applied to the complete signal when the speech enhancement is processed using segmental speech enhancement. The enhanced signal is further improved by applying a soft wavelet thresholding technique to the un-segmented enhanced speech at the final processing stage. The results show improvements both quantitatively and qualitatively compared to the speech enhancement that does not apply the CFMA algorithm.

Dariush Farrokhi - One of the best experts on this subject based on the ideXlab platform.

  • Single channel speech enhancement using a 9 Dimensional Noise Estimation algorithm and Controlled Forward March Averaging
    2008 9th International Conference on Signal Processing, 2008
    Co-Authors: Dariush Farrokhi, Roberto Togneri, Anthony Zaknich
    Abstract:

    A post processing technique is proposed to enhance speech in a single channel system. A new noise estimation algorithm is proposed in conjunction with the Controlled Forward March Averaging (CFMA) technique to enhance speech in a single channel non-stationary noisy system. We introduce a 9-Dimensional Noise Estimation (NDNE) algorithm to the Single Channel Speech Estimation (SCSE) system, that updates the estimated noise in 9 frequency sub-bands, by averaging the noisy speech power spectrum using a time and frequency dependent smoothing Factor. A signal presence Probability Factor is calculated by computing the ratio of the noisy speech power spectrum to its local minimum, which is computed by averaging past values of the noisy speech power spectra with a look-ahead Factor. The NDNE uses a non-linear thresholding map as oppose to the conventional linear thresholding. This new algorithm produced an average 7% improvement in 0 and -2.5 dB global SNR in speech corrupted with modified Babble noise. Subjective tests confirmed these results.

  • Speech enhancement of non-stationary noise based on controlled forward moving average
    2007 International Symposium on Communications and Information Technologies, 2007
    Co-Authors: Dariush Farrokhi, Roberto Togneri, Anthony Zaknich
    Abstract:

    A pre and post processing technique is proposed to enhance the speech signal of highly non-stationary noisy speech. The purpose of this research has been to build on current speech enhancement algorithms to produce an improved algorithm for enhancement of speech contaminated with non-stationary babble type noise. The pre processing involves two stages. In stage one, the variance of the noisy speech spectrum is reduced by utilizing the Discrete or Prolate Spheroidal Sequence (DPSS) multi-taper algorithm plus a Controlled Forward Moving Average (CFMA) technique. We introduced the CFMA algorithm to smooth and reduce variance of the estimated non-stationary noise spectrum. In the second stage the noisy speech power spectrum is de-noised by applying Stein's Unbiased Risk Estimator (SURE) wavelet thresholding technique. In the third layer, use is made of a noise estimation algorithm with rapid adaptation for a highly non-stationary noise environment. The noise estimate is updated in three frequency sub-bands, by averaging the noisy speech power spectrum using a frequency dependent smoothing Factor, which is adjusted, based on a signal presence Probability Factor. In the fourth layer a spectral subtraction algorithm is used to enhance the speech signal, by subtracting each estimated noise from the original noisy speech. The new proposed post processing is then applied to the complete signal when the speech enhancement is processed using segmental speech enhancement. The enhanced signal is further improved by applying a soft wavelet thresholding technique to the un-segmented enhanced speech at the final processing stage. The results show improvements both quantitatively and qualitatively compared to the speech enhancement that does not apply the CFMA algorithm.

Roberto Togneri - One of the best experts on this subject based on the ideXlab platform.

  • Single channel speech enhancement using a 9 Dimensional Noise Estimation algorithm and Controlled Forward March Averaging
    2008 9th International Conference on Signal Processing, 2008
    Co-Authors: Dariush Farrokhi, Roberto Togneri, Anthony Zaknich
    Abstract:

    A post processing technique is proposed to enhance speech in a single channel system. A new noise estimation algorithm is proposed in conjunction with the Controlled Forward March Averaging (CFMA) technique to enhance speech in a single channel non-stationary noisy system. We introduce a 9-Dimensional Noise Estimation (NDNE) algorithm to the Single Channel Speech Estimation (SCSE) system, that updates the estimated noise in 9 frequency sub-bands, by averaging the noisy speech power spectrum using a time and frequency dependent smoothing Factor. A signal presence Probability Factor is calculated by computing the ratio of the noisy speech power spectrum to its local minimum, which is computed by averaging past values of the noisy speech power spectra with a look-ahead Factor. The NDNE uses a non-linear thresholding map as oppose to the conventional linear thresholding. This new algorithm produced an average 7% improvement in 0 and -2.5 dB global SNR in speech corrupted with modified Babble noise. Subjective tests confirmed these results.

  • Speech enhancement of non-stationary noise based on controlled forward moving average
    2007 International Symposium on Communications and Information Technologies, 2007
    Co-Authors: Dariush Farrokhi, Roberto Togneri, Anthony Zaknich
    Abstract:

    A pre and post processing technique is proposed to enhance the speech signal of highly non-stationary noisy speech. The purpose of this research has been to build on current speech enhancement algorithms to produce an improved algorithm for enhancement of speech contaminated with non-stationary babble type noise. The pre processing involves two stages. In stage one, the variance of the noisy speech spectrum is reduced by utilizing the Discrete or Prolate Spheroidal Sequence (DPSS) multi-taper algorithm plus a Controlled Forward Moving Average (CFMA) technique. We introduced the CFMA algorithm to smooth and reduce variance of the estimated non-stationary noise spectrum. In the second stage the noisy speech power spectrum is de-noised by applying Stein's Unbiased Risk Estimator (SURE) wavelet thresholding technique. In the third layer, use is made of a noise estimation algorithm with rapid adaptation for a highly non-stationary noise environment. The noise estimate is updated in three frequency sub-bands, by averaging the noisy speech power spectrum using a frequency dependent smoothing Factor, which is adjusted, based on a signal presence Probability Factor. In the fourth layer a spectral subtraction algorithm is used to enhance the speech signal, by subtracting each estimated noise from the original noisy speech. The new proposed post processing is then applied to the complete signal when the speech enhancement is processed using segmental speech enhancement. The enhanced signal is further improved by applying a soft wavelet thresholding technique to the un-segmented enhanced speech at the final processing stage. The results show improvements both quantitatively and qualitatively compared to the speech enhancement that does not apply the CFMA algorithm.

Haibo Hu - One of the best experts on this subject based on the ideXlab platform.

  • A numerical study of multiple adiabatic shear bands evolution in a 304LSS thick-walled cylinder
    2017
    Co-Authors: Haibo Hu, Tiegang Tang
    Abstract:

    The self-organization of multiple shear bands in a 304L stainless steel(304LSS) thick-walled cylinder (TWC) was numerically studied. The microstructures of material lead to the non-uniform distribution of the local yield stress, which play a key role in the formation of spontaneous shear localization. We introduced a Probability Factor satisfied the Gaussian distribution into the macroscopic constitutive relationship to describe the non-uniformity of local yield stress. Using the Probability Factor, the initiation and propagation of multiple shear bands in TWC were numerically replicated in our 2D FEM simulation. Experimental results in the literature indicated that the machined surface at the internal boundary of a 304L stainless steel cylinder provides a work-hardened layer (about 20∼30μm) which has significantly different microstructures from the base material. The work-hardened layer leads to the phenomenon that most shear bands propagate along a given direction, clockwise or counterclockwise. In our ...

  • A numerical model for adiabatic shear bands with application to a thick-walled cylinder in 304 stainless steel
    Modelling and Simulation in Materials Science and Engineering, 2013
    Co-Authors: Yongchi Li, Haibo Hu, Xiuzhang Hu
    Abstract:

    The formation of an adiabatic shear band (ASB) experiences three stages: stable plastic flow, nucleation and a fluid-like stage. For different stages, the microstructures of the material undergo great changes. The mechanical behavior of the material in each stage has its own unique characteristics. To describe these characteristics, a multi-stage model for the shear band is proposed. For the stable plastic flow stage, a modified adiabatic J–C constitutive relationship is used. For the nucleation stage, the effects of work hardening and temperature softening are described by a power function of plastic strain. A Newtonian fluid model is used for the fluid-like stage. The formation of a shear band is an instability process. Various defects in the material are perturbation sources, which change the local yield stress. To describe the disturbances, a Probability Factor is introduced into the macroscopic constitutive relationship. The yield stress in the material is assumed to obey a Gaussian distribution. The multi-stage model combined with a Probability Factor is applied to simulate the rupture of thick-walled cylinder in 304 Stainless Steel (304SS). A close agreement is found between the simulation and experimental results, such as the failure mechanism, shear band spacing and propagating velocity of the shear band. By combining the experimental results with the simulation results, the importance of the nucleation stage is emphasized.

Tiegang Tang - One of the best experts on this subject based on the ideXlab platform.

  • A numerical study of multiple adiabatic shear bands evolution in a 304LSS thick-walled cylinder
    2017
    Co-Authors: Haibo Hu, Tiegang Tang
    Abstract:

    The self-organization of multiple shear bands in a 304L stainless steel(304LSS) thick-walled cylinder (TWC) was numerically studied. The microstructures of material lead to the non-uniform distribution of the local yield stress, which play a key role in the formation of spontaneous shear localization. We introduced a Probability Factor satisfied the Gaussian distribution into the macroscopic constitutive relationship to describe the non-uniformity of local yield stress. Using the Probability Factor, the initiation and propagation of multiple shear bands in TWC were numerically replicated in our 2D FEM simulation. Experimental results in the literature indicated that the machined surface at the internal boundary of a 304L stainless steel cylinder provides a work-hardened layer (about 20∼30μm) which has significantly different microstructures from the base material. The work-hardened layer leads to the phenomenon that most shear bands propagate along a given direction, clockwise or counterclockwise. In our ...