Development of a Real-Time Audio Signal Processing System for Speech Enhancement

Abstract

The requirements of real-time signal processing dictate that the audio signal must be completely processed before the subsequent audio segment may be received. This is done in order to achieve those requirements. This highlights how important it is to create methods of signal processing that are not just quick but also accurate. I describe many ways for processing audio signals in real time within the scope of this thesis. The publications that are being presented cover a wide range of issues, including noise dosimetry, speech analysis, and network echo cancellation, to name a few. In this article, the process of constructing a system that uses audio signal processing to improve speech in real time is broken down and examined. Speech enhancement makes speech more audible and comprehensible in noisy surroundings, which is beneficial to individuals who have hearing loss as well as speech recognition and communication systems. Signal-to-noise ratio, power spectral efficiency ratio, and spectrum transfer entropy index are the metrics that are utilized to evaluate speech quality enhancement system development strategies, abbreviated as STOI. Because studies have shown that being exposed to loud noises over extended periods of time can have severe impacts on health, it is essential to have precise methods for detecting the levels of noise. The findings of a study that measured exposure to noise while also taking into account the impact of the speaker's own voice are presented in this article.