Pole-zero modeling of speech based on high-order pole model fitting and decomposition method

In this paper four pole-zero modeling algorithms of clean and noisy speech have been studied in a unified approach that is based on high-order pole model fitting and decomposition method. They are autocorrelation prediction (AP), modified Yule-Walker (MYW), modified least square (MLS), and modified least square with autocorrelation compensation (MLSAC) methods. They involve only linear equations, and therefore are computationally efficient. Among these algorithms, the MLSAC method appears to be the most effective in spectral envelope estimation of noisy as well as clean speech. According to our simulation results, the improvement resulting from the use of the MLSAC pole-zero model for noisy speech is equivalent to increasing signal-to-noise ratio (SNR) by about 5 dB when SNR of input speech is 10 dB or less. The use of a pole-zero model in multirate vocoding is also discussed.