Probabilistic Modeling of Speech in Spectral Domain using Maximum Likelihood Estimation

Abstract

The performance of many speech processing algorithms depends on modeling speech signals using appropriate probability distributions. Various distributions such as the Gamma distribution, Gaussian distribution, Generalized Gaussian distribution, Laplace distribution as well as multivariate Gaussian and Laplace distributions have been proposed in the literature to model different segment lengths of speech, typically below 200 ms in different domains. In this paper, we attempted to fit Laplace and Gaussian distributions to obtain a statistical model of speech short-time Fourier transform coefficients with high spectral resolution (segment length >500 ms) and low spectral resolution (segment length <10 ms). Distribution fitting of Laplace and Gaussian distributions was performed using maximum-likelihood estimation. It was found that speech short-time Fourier transform coefficients with high spectral resolution can be modeled using Laplace distribution. For low spectral resolution, neither the Laplace nor Gaussian distribution provided a good fit. Spectral domain modeling of speech with different depths of spectral resolution is useful in understanding the perceptual stability of hearing which is necessary for the design of digital hearing aids.

Dataset of Raw and Pre-processed Speech Signals, Mel Frequency Cepstral Coefficients of Speech and Heart Rate Measurements

“…Thus the resulting audio bit rate is R b = nf S = 256 kbps. These parameters are adequate to faithfully represent speech signals [6], [7].These settings are configurable in the Matlab script which also activates the microphone. All instructions are provided to the volunteers who are made to rest for 30 minutes before recording their data in order to collect data corresponding to resting heart rate.…”

The Theory behind Controllable Expressive Speech Synthesis: A Cross-Disciplinary Approach

“…To choose the right criterium to optimize when working with speech data, one should pay attention to speech probability distributions. Speech waveforms and magnitude spectrogram distribution are Laplacian [21,22]. That is why MAE loss should be used to optimize their predictions.…”

Significance of Intelligent Pause Detection Protocol (IPDP) Over Other Protocols used for Speech Processing

“…Here in this paper the mean value and the variance are estimated using MLE Algorithm. M.Usman,etl [1] observed performance for speech processing algorithms for various distribution functions by omitting pauses. J.…”