Artificial bandwidth extension using deep neural network and $H^\infty$ sampled-data control theory

TL;DR

This paper introduces a novel approach combining $H^ Infty$ sampled-data control theory and deep neural networks to enhance speech bandwidth, effectively recovering high-frequency components for improved telephonic speech quality.

Contribution

The paper presents a new method that integrates sampled-data control theory with deep neural networks for artificial bandwidth extension of speech signals.

Findings

Effective high-frequency recovery demonstrated on TIMIT and RSR15 datasets.

Improved speech quality in both voiced and unvoiced segments.

Objective and subjective evaluations confirm the method's efficacy.

Abstract

Artificial bandwidth extension is applied to speech signals to improve their quality in narrowband telephonic communication. For accomplishing this, the missing high-frequency (high-band) components of speech signals are recovered by utilizing a new extrapolation process based on sampled-data control theory and deep neural network (DNN). The $H^\infty$ sampled-data control theory helps in designing of a high-band filter to recover the high-frequency signals by optimally utilizing the inter-sample signals. Non-stationary (time-varying) characteristics of speech signals forces to use numerous high-band filters. Hence, we use a deep neural network for estimating the high-band filter information and a gain factor for a specified narrowband information of the unseen signal. The objective analysis is done on the TIMIT dataset and RSR15 dataset. Additionally, the objective analysis is performed separately for the voiced speech as well as for the unvoiced speech as generally needed in speech processing. Subjective analysis is done on the RSR15 dataset.