Extraction and analysis of voice samples based on short audio files

Abstract

Some voice defects may be removed from a recording in such a way that the remainder of the sample contains only the relevant data. However processing of input sounds to achieve this goal requires methodology that will be able to distinguish between clear sound and noise. In this paper, we propose data extraction technique and defect analysis. Developed methodology may allow for possible use in various life support systems, where sounds are processed to verify identity, control, communicate, etc. Proposed method has been tested and compared to other indirect methods.