Development of a System for Assessing the Dynamics of the Lecturer's Speech Expressiveness Indicators During the Lecture

This document considers the study of the effectiveness of speech analysis methods in Python desktop application for learning classes. In this study, an application for analyzing audio files by various speech and fundamental frequency analysis methods was developed. Classification of speech and fundamental frequency analysis methods was completed, as well as a comparison of the efficiency of these methods with the prepared dataset. These methods were implemented in the application. Neural networks were developed and implemented in the application for the analysis of emotional and expressive characteristics of speech. The objective of this study is to develop an application for analyzing the user’s speech by various methods. The result of this study is an application that allows you to analyze speech in real-time and in the form of an audiofile using trained neural networks. The result also includes an analysis of the effectiveness of methods for analyzing fundamental frequency, carried out using the developed application. This document contains 85 pages, 13 illustrations, 2 tables and 31 references.