Spectral Zero-Crossings: Localization Properties and Applications

Shenoy, Ravi R.; Seelamantula, Chandra Sekhar

doi:10.1109/tsp.2015.2420538

Cited by 15 publications

(7 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The process is approximately stationary Gaussian when the number of sinusoids is large [4]. Although this behaviour is asymptotic, we have shown in [30] that the Gaussian assumption is accurate even for moderate values of . The ZCR of at lag is related to its autocorrelation as .…”

Section: Application To Robust Autocorrelation Estimationmentioning

confidence: 93%

A Zero-Crossing Rate Property of Power Complementary Analysis Filterbank Outputs

Shenoy

Seelamantula

2015

IEEE Signal Process. Lett.

Self Cite

View full text Add to dashboard Cite

We establish zero-crossing rate (ZCR) relations between the input and the subbands of a maximally decimated -channel power complementary analysis filterbank when the input is a stationary Gaussian process. The ZCR at lag is defined as the number of sign changes between the samples of a sequence and its -sample shifted version, normalized by the sequence length. We derive the relationship between the ZCR of the Gaussian process at lags that are integer multiples of and the subband ZCRs. Based on this result, we propose a robust iterative autocorrelation estimator for a signal consisting of a sum of sinusoids of fixed amplitudes and uniformly distributed random phases. Simulation results show that the performance of the proposed estimator is better than the sample autocorrelation over the SNR range of to 15 dB. Validation on a segment of a trumpet signal showed similar performance gains.

show abstract

Section: Application To Robust Autocorrelation Estimationmentioning

confidence: 93%

A Zero-Crossing Rate Property of Power Complementary Analysis Filterbank Outputs

Shenoy

Seelamantula

2015

IEEE Signal Process. Lett.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore, to improve the recognition rate of baby cry, hybrid features combining the multi-domain of baby cry is proposed. This hybrid feature includes frequency domain feature MFCC, Log-Mel Spectrogram [29], and time domain feature Zero Crossing Rate (ZCR) [30].…”

Section: ) Hybrid Featurementioning

confidence: 99%

Baby Cry Recognition by BCRNet Using Transfer Learning and Deep Feature Fusion

Zhang,

Ting,

Choo

2023

IEEE Access

View full text Add to dashboard Cite

Deep learning theory has made remarkable advancements in the field of baby cry recognition, significantly enhancing its accuracy. Nonetheless, existing research faces two challenges. Firstly, the limited size of the database increases the risk of overfitting for a deep learning model. Secondly, the integration of multi-domain features has been neglected. To address these issues, a novel approach called BCRNet is proposed, which combines transfer learning and feature fusion. The BCRNet model takes multi-domain features as input and extracts deep features using a transfer learning model. Subsequently, a multilayer autoencoder is utilized for feature reduction, and a Support Vector Machine (SVM) is employed to select the transfer learning model with the highest classification accuracy. Then two features are concatenated to form fused features. Finally, the fused features are fed into a deep neural network for classification. Experimental results show that the proposed model is effective in mitigating the model overfitting problem due to small datasets. The fused features of the proposed method are better than the existing methods using single domain feature.

show abstract

“…In the case of voiced sounds, the impulses are placed at the instants of significant excitation (referred to as epochs or glottal closure instants (GCIs)) [17,41,51], which depends on the pitch of the speaker. The vocal-tract configuration in producing a certain sound determines the frequency response of the filter.…”

Section: While Flag False Domentioning

confidence: 99%

A Non-Convex Optimization Technique for Sparse Blind Deconvolution -- Initialization Aspects and Error Reduction Properties

Adiga,

Seelamantula

2017

Preprint

Self Cite

View full text Add to dashboard Cite

Sparse blind deconvolution is the problem of estimating the blur kernel and sparse excitation, both of which are unknown. Considering a linear convolution model, as opposed to the standard circular convolution model, we derive a sufficient condition for stable deconvolution. The columns of the linear convolution matrix form a Riesz basis with the tightness of the Riesz bounds determined by the autocorrelation of the blur kernel. Employing a Bayesian framework results in a non-convex, non-smooth cost function consisting of an 2 data-fidelity term and a sparsity promoting p-norm (0 ≤ p ≤ 1) regularizer. Since the p-norm is not differentiable at the origin, we employ an -regularized p-norm as a surrogate. The data term is also non-convex in both the blur kernel and excitation. An iterative scheme termed alternating minimization (Alt. Min.) p − 2 projections algorithm (ALPA) is developed for optimization of the -regularized cost function. Further, we demonstrate that, in every iteration, the -regularized cost function is non-increasing and more importantly, bounds the original p-norm-based cost. Due to non-convexity of the cost, the accuracy of estimation is largely influenced by the initialization. Considering regularized least-squares estimate as the initialization, we analyze how the initialization errors are concentrated, first in Gaussian noise, and then in bounded noise, the latter case resulting in tighter bounds. Comparisons with state-of-the-art blind deconvolution algorithms show that the deconvolution accuracy is higher in case of ALPA. In the context of natural speech signals, ALPA results in accurate deconvolution of a voiced speech segment into a sparse excitation and smooth vocal tract response.

show abstract

Spectral Zero-Crossings: Localization Properties and Applications

Cited by 15 publications

References 40 publications

A Zero-Crossing Rate Property of Power Complementary Analysis Filterbank Outputs

A Zero-Crossing Rate Property of Power Complementary Analysis Filterbank Outputs

Baby Cry Recognition by BCRNet Using Transfer Learning and Deep Feature Fusion

A Non-Convex Optimization Technique for Sparse Blind Deconvolution -- Initialization Aspects and Error Reduction Properties

Contact Info

Product

Resources

About