Mitigating Information Interruptions by COVID-19 Face Masks: A Three-Stage Speech Enhancement Scheme

The coronavirus disease 2019 (COVID-19) preventive measures have resulted in significant lifestyle changes. One of the COVID-19 new normal is the usage of face masks for protection against airborne aerosol which creates distractions and interruptions in voice communication. It has a different influence on speech than the standard concept of noise affecting speech communication. Furthermore, it has varied effects on speech in different frequency bands. To provide a solution to this problem, a three-stage adaptive speech enhancement (SE) scheme is developed in this article. In the first stage, the tunable $Q$ -factor wavelet transform (TQWT) features are extracted by properly setting the quality factor values and the number of levels from the input speech signal. In the second stage, the adjustable parameters of the preemphasis filter and modified multiband spectral subtraction (MBSS) are determined using bio-inspired techniques for different masking and signal-to-noise ratio (SNR) conditions. In the third stage, the weights, center values, standard deviation of the Gaussian radial basis functions, and input patterns of the radial basis function neural networks (RBFNNs) are updated to predict the optimized parameters from the input TQWT-based cepstral features (TQCFs). In the end, the performance of the proposed algorithm is compared with the standard SE algorithms using two speech datasets.

I. Introduction

The coronavirus disease 2019 (COVID-19) started in December 2019 and it has become a pandemic causing more than six million deaths across the globe [1], [2]. Face masks have been recommended by healthcare officials around the world as a precautionary measure to prevent the spread of this virus [3], [4]. It has been observed that there are different varieties of masks being used, and they have significantly different effects on speech intelligibility [5]. Face masks cause speech distortion and have a detrimental impact on speech perception, lowering the quality and intelligibility of speech [6], [7]. A comprehensive review of the impact of wearing a mask on speech production is reported in [8]. Considerable attenuation of signal strength at high frequencies between 2 and 16 kHz for all the mask designs is observed, resulting in an increased vocal effort, discomfort, and complications. The influence of masks on sound transmission in a lecture hall has been investigated in [9]. Male vocals are more affected by the use of masks than female voices. The signal has been lowered by 8–16 dB in the frequency range of 1.6–6.3 kHz. The use of face masks has raised the prevalence of vocal issues among healthcare workers [10]. Another research reveals that using surgical masks and face shields has an adverse impact on people with moderate to severe hearing loss [11]. Face masks have been observed to cause frequency-dependent transmission loss. When compared with testing without a mask, the directivity index deviates by up to 7 dB for frequencies over 3 kHz [12].