Cepstrum

About: Cepstrum is a research topic. Over the lifetime, 3346 publications have been published within this topic receiving 55742 citations.

Engine Misfire Detection with Pervasive Mobile Audio

Abstract: We address the problem of detecting whether an engine is misfiring by using machine learning techniques on transformed audio data collected from a smartphone. We recorded audio samples in an uncontrolled environment and extracted Fourier, Wavelet and Mel-frequency Cepstrum features from normal and abnormal engines. We then implemented Fisher Score and Relief Score based variable ranking to obtain an informative reduced feature set for training and testing classification algorithms. Using this feature set, we were able to obtain a model accuracy of over 99 % using a linear SVM applied to outsample data. This application of machine learning to vehicle subsystem monitoring simplifies traditional engine diagnostics, aiding vehicle owners in the maintenance process and opening up new avenues for pervasive mobile sensing and automotive diagnostics.

Feature Transformations and Combinations for Improving ASR Performance

Abstract: In this work, linear and nonlinear feature transformations have been experimented in ASR front end. Unsupervised transformations were based on principal component analysis and independent component analysis. Discriminative transformations were based on linear discriminant analysis and multilayer perceptron networks. The acoustic models were trained using a subset of HUB5 training data and they were tested using OGI Numbers corpus. Baseline feature vector consisted of PLP cepstrum and energy with first and second order deltas. None of the feature transformations could outperform the baseline when used alone, but improvement in the word error rate was gained when the baseline feature was combined with the feature transformation stream. Two combination methods were experimented: feature vector concatenation and n-best list combination using ROVER. Best results were obtained using the combination of the baseline PLP cepstrum and the feature transform based on multilayer perceptron network. The word error rate in the number recognition task was reduced from 4.1 to 3.1.

Voice source parameters for speaker verification

Abstract: In this paper we report on a study of the variability of voice source parameters in the context of speaker characterisation, and we propose a speaker verification system which incorporates these parameters. The motivation for this approach is that, whilst we have conscious control over the action of our vocal tract articulators such as the tongue and jaw, we have only limited voluntary muscle control over the vocal cords. The conjecture is, therefore, that impostors are less likely to be able to mimic vocal cord effects than vocal tract effects. The hybrid speaker verification system that is proposed incorporates two sub-systems to improve the overall performance: (i) a cepstral-based HMM with cohort normalisation and (ii) voice source parameters derived from Multi-cycle Closed-phase Glottal Inverse Filtering (MCGIF). Preliminary experimental results show that the hybrid system performs better than either of the sub-systems in terms of the equal error rate (EER). Specifically, the hybrid system improved the performance of the cepstral-based HMM system by 78% on average, resulting in a mean EER of 0.42% for the specific tests conducted.

Exact Phase Retrieval for a Class of 2-D Parametric Signals

Abstract: We address the problem of two-dimensional (2-D) phase retrieval from magnitude of the Fourier spectrum. We consider 2-D signals that are characterized by first-order difference equations, which have a parametric representation in the Fourier domain. We show that, under appropriate stability conditions, such signals can be reconstructed uniquely from the Fourier transform magnitude. We formulate the phase retrieval problem as one of computing the parameters that uniquely determine the signal. We show that the problem can be solved by employing the annihilating filter method, particularly for the case when the parameters are distinct. For the more general case of the repeating parameters, the annihilating filter method is not applicable. We circumvent the problem by employing the algebraically coupled matrix pencil (ACMP) method. In the noiseless measurement setup, exact phase retrieval is possible. We also establish a link between the proposed analysis and 2-D cepstrum. In the noisy case, we derive Cramer–Rao lower bounds (CRLBs) on the estimates of the parameters and present Monte Carlo performance analysis as a function of the noise level. Comparisons with state-of-the-art techniques in terms of signal reconstruction accuracy show that the proposed technique outperforms the Fienup and relaxed averaged alternating reflections (RAAR) algorithms in the presence of noise.

Single shot three-dimensional imaging using an engineered point spread function.

Abstract: A system approach to acquire a three-dimensional object distribution is presented using a compact and cost efficient camera system with an engineered point spread function. The corresponding monocular setup incorporates a phase-only computer-generated hologram in combination with a conventional imaging objective in order to optically encode the axial information within a single two-dimensional image. The object’s depth map is calculated using a novel approach based on the power cepstrum of the image. The in-plane RGB image information is restored with an extended depth of focus by applying an adapted Wiener filter. The presented approach is tested experimentally by estimating the three-dimensional distribution of an extended passively illuminated scene.