Audio Signal Preparation Process for Deep Learning Application Using Python

This paper was created as a part of the project "Development of software to improve communication, academic and social skills of children with disabilities. Artificial Intelligence today represents a wide area of different computer algorithms and systems with one main goal – to mimic and eventually replace human thinking and logic. One of the most important parts of the mentioned software in development is correct and meaningful collection and preparation of relevant data. Deep learning model structure, model training and results heavily depend on thoughtful identification and processing of relevant data. As a part of a wider project, this paper is representing a short overview of sound record digitalization and recommended steps required in data preparation for use in artificial intelligence applications.