Showing papers by "Goutam Saha published in 2007"
TL;DR: The work presented here needs only the average heart rate as discrete auxiliary information that can be easily provided, unlike most of the methods which require the electrocardiography (ECG) signal as a continuous auxiliary signal in a complex setup.
Abstract: The detection of heart diseases from heart sound signals needs an efficient segmentation algorithm to properly identify the location of the first and second heart sounds. This in turn helps in characterizing murmurs present in the cardiac cycles and the pathological condition by providing an appropriate time reference. The work presented here needs only the average heart rate as discrete auxiliary information that can be easily provided, unlike most of the methods which require the electrocardiography (ECG) signal as a continuous auxiliary signal in a complex setup. The algorithm was tested on 34 pathological cases and normal heart sound for a variety of sampling frequencies, recording environments, and age groups of subjects. It was found to give an overall accuracy of 95.51%. The robustness of the algorithm against additive white Gaussian noise contamination is also presented, and the noise immunity of various diseases for correct segmentation is established through this study.
27 citations
23 Sep 2007
TL;DR: Experimental results show that the proposed estimator yields a higher improvement in identification accuracy compared to other estimators on popular Gaussian Mixture Model (GMM) based speaker model and Mel-Frequency Cepstral Coefficient (MFCC) features.
Abstract: Real world Speaker Identification (SI) application differs from ideal or laboratory conditions causing perturbations that leads to a mismatch between the training and testing environment and degrade the performance drastically. Many strategies have been adopted to cope with acoustical degradation; wavelet based Bayesian marginal model is one of them. But Bayesian marginal models cannot model the inter-scale statistical dependencies of different wavelet scales. Simple nonlinear estimators for wavelet based denoising assume that the wavelet coefficients in different scales are independent in nature. However wavelet coefficients have significant inter-scale dependency. This paper enhances this inter-scale dependency property by a Circularly Symmetric Probability Density Function (CS-PDF) related to the family of Spherically Invariant Random Processes (SIRPs) in Log Gabor Wavelet (LGW) domain and corresponding joint shrinkage estimator is derived by Maximum a Posteriori (MAP) estimator. A framework is proposed based on these to denoise speech signal for automatic speaker identification problems. The robustness of the proposed framework is tested for Text Independent Speaker Identification application on 100 speakers of POLYCOST and 100 speakers of YOHO speech database in three different noise environments. Experimental results show that the proposed estimator yields a higher improvement in identification accuracy compared to other estimators on popular Gaussian Mixture Model (GMM) based speaker model and Mel-Frequency Cepstral Coefficient (MFCC) features. Keywords—Speaker Identification, Log Gabor Wavelet, Bayesian Bivariate Estimator, Circularly Symmetric Probability Density Function, SIRP.
14 citations