Approximate Kalman Filtering for the Harmonic plus Noise Model
نویسندگان
چکیده
We present a probabilistic description of the Harmonic plus Noise Model (HNM) for speech signals. This probabilistic formulation permits Maximum Likelihood (ML) parameter estimation and speech synthesis becomes a straightforward sampling from a distribution. It also permits development of a Kalman filter that tracks model parameters such as pitch, harmonic amplitudes, and autoregressive coefficients. We focus here on pitch tracking for which the estimator is highly non-linear. As a result it is necessary to develop an approximate Kalman filter that goes beyond extended Kalman filtering. 1. THE HARMONIC PLUS NOISE MODEL Since the work of McAulay and Quatieri [1] speech has been repeatedly modeled as the sum of harmonic sinusoids in additive noise. Based on this model speech synthesis and morphing with high perceptual quality has been achieved among other applications (see references in [2, 3]). In the Harmonic plus Noise Model (HNM) the observed data y(t) is the sum of a harmonic component h(t), which captures the voiced portion of the speech spectrum while a colored noise component, n(t) captures the unvoiced portion of speech y(t) = h(t) + n(t) : (1) The harmonic and noise processes are defined as
منابع مشابه
Application of the Kalman-Bucy filter in the stochastic differential equation for the modeling of RL circuit
In this paper, we present an application of the stochastic calculusto the problem of modeling electrical networks. The filtering problem have animportant role in the theory of stochastic differential equations(SDEs). In thisarticle, we present an application of the continuous Kalman-Bucy filter for a RLcircuit. The deterministic model of the circuit is replaced by a stochastic model byadding a ...
متن کاملGPS Time-Variable Seasonal Signals Modeling Master Thesis
Seasonal signals (annual plus semi-annual) in GPS time series are of great importance for understanding the evolution of regional mass, i.e. ice and hydrology. Conventionally these signals (annual and semi-annual) are derived by least-squares fitting of harmonic terms with a constant amplitude and phase. In reality, however, such seasonal signals are modulated, i.e. they will have a time-variab...
متن کاملA Kalman filtering approach to short-time Fourier analysis
The problem of estimating time-varying harmonic components of a signal measured in noise is considered. The approach used is via state estimation. Two methods are proposed, one involving poleplacement of a state observer, the other using quadratic optimization techniques. The result is the development of a new class of filters, akin to recursive frequency-sampling filters, for inclusion in a pa...
متن کاملA FILTERED B-SPLINE MODEL OF SCANNED DIGITAL IMAGES
We present an approach for modeling and filtering digitally scanned images. The digital contour of an image is segmented to identify the linear segments, the nonlinear segments and critical corners. The nonlinear segments are modeled by B-splines. To remove the contour noise, we propose a weighted least q m s model to account for both the fitness of the splines as well as their approximate cur...
متن کاملSigma-Point Filtering and Smoothing Based Parameter Estimation in Nonlinear Dynamic Systems
We consider approximate maximum likelihood parameter estimation in nonlinear state-space models. We discuss both direct optimization of the likelihood and expectation– maximization (EM). For EM, we also give closed-form expressions for the maximization step in a class of models that are linear in parameters and have additive noise. To obtain approximations to the filtering and smoothing distrib...
متن کامل