Speech Decomposition Based on a Hybrid Speech Model and Optimal Segmentation

TL;DR

This paper introduces an adaptive segmentation method for decomposing noisy speech into voiced and unvoiced components using a hybrid speech model, improving extraction accuracy over fixed segmentation approaches.

Contribution

It proposes a novel adaptive segmentation technique based on MAP and log-likelihood criteria, enhancing speech component separation in noisy environments.

Findings

Improved component extraction accuracy with adaptive segmentation.

Lower distortion in voiced speech compared to fixed segmentation methods.

Higher segmental SNR for both components than existing approaches.

Abstract

In a hybrid speech model, both voiced and unvoiced components can coexist in a segment. Often, the voiced speech is regarded as the deterministic component, and the unvoiced speech and additive noise are the stochastic components. Typically, the speech signal is considered stationary within fixed segments of 20-40 ms, but the degree of stationarity varies over time. For decomposing noisy speech into its voiced and unvoiced components, a fixed segmentation may be too crude, and we here propose to adapt the segment length according to the signal local characteristics. The segmentation relies on parameter estimates of a hybrid speech model and the maximum a posteriori (MAP) and log-likelihood criteria as rules for model selection among the possible segment lengths, for voiced and unvoiced speech, respectively. Given the optimal segmentation markers and the estimated statistics, both components are estimated using linear filtering. A codebook-based approach differentiates between unvoiced speech and noise. A better extraction of the components is possible by taking into account the adaptive segmentation, compared to a fixed one. Also, a lower distortion for voiced speech and higher segSNR for both components is possible, as compared to other decomposition methods.