Development of a hardware and software complex for speech analysis and correction

Abstract

In recent years, there has been a dramatic increase in the number of people suffering from functional disorders of voice, usually caused by a psychoemotional stress. Such disorders bring significant discomfort to a person's life as they reduce their communication and social adaptation capacitty, which in turn increases the psychoemotional load. As a result, functional disorders are fixed by the vicious circle mechanism o and can be transformed into the pathology of the speech apparatus. The main method of diagnosis remains expert assessment, which directly depends on the professional skills of a specialist in working with voice. In this connection, the issue of developing such systems for diagnosing voice-speech disorders that would allow for an objective assessment based on the processing of voice-speech characteristics, as well as to identify the violation in time and prevent the development of pathology, is relevant. Such methods and systems can be useful both for diagnostics and for monitoring the effectiveness of voice therapy. The existing methods of hardware diagnostics have not yet found their application in practice due to their inconsistency with the results of expert evaluation. In this paper, we propose a new concept of hardware and software complex for the analysis of voice based on acoustic characteristics of a set of harmonics of the voice signal. A VASA (Voice and Speech Analyzing system) complex has been developed that provides an automatic analysis of the amplitudes of the first 16 harmonics. The tests performed on three volunteers showed a high level of reproducibility and repeatability (within 10 % < %R&R < 30 %), sufficient for conducting comparative studies on healthy people and people with functional speech disorders.