The analysis of signals that can be represented as a linear combination of exponentially damped sinusoids where the values of damping factors, frequencies, and the linear combination coefficients change at certain transition times is considered. These transitions represent the opening and closing of the glottis in the case of speech signals. Techniques are presented for the accurate estimation of the exponential parameters and the times of transition, from noise corrupted observations of the signal. The exponential parameters are obtained by improved linear prediction techniques using low-rank approximations, and further refined by an iterative least-squares technique with stability constraints imposed on the damping factors. Optimal estimates (in the least-squares sense) of the time of transition are presented. Our knowledge of the signal structure is used to obtain improved performance and also a computationally efficient estimation algorithm. Experiments with real, connected speech indicate that the speech waveforms can be accurately represented from a small number of parameters using the analysis presented here.
[1]
A. Evans,et al.
Optimal least squares time-domain synthesis of recursive digital filters
,
1973
.
[2]
Donald W. Tufts,et al.
Excitation-synchronous modeling of voiced speech
,
1987,
IEEE Trans. Acoust. Speech Signal Process..
[3]
R. Kumaresan,et al.
Estimating the parameters of exponentially damped sinusoids and pole-zero modeling in noise
,
1982
.
[4]
L. Scharf,et al.
A Prony method for noisy data: Choosing the signal components and selecting the order in exponential signal models
,
1984,
Proceedings of the IEEE.
[5]
S. Parthasarathy,et al.
Maximum-likelihood estimation of parameters of exponentially damped sinusoids
,
1985,
Proceedings of the IEEE.
[6]
R. Kumaresan,et al.
Estimation of frequencies of multiple sinusoids: Making linear prediction perform like maximum likelihood
,
1982,
Proceedings of the IEEE.