Using Formants for Human Speech Recognition by Artificial Intelligence

Abstract

The global transition to the digital space and digital transformation form the needs and challenges not only in the creation of new digital devices, but also in the translation of business processes (for example, Data Governance), which allows a significant reduction of transaction costs and a significant increase in the volume of economic activity. Today, many large companies from many areas of production use a voice assistant in their services, thus reducing the burden on operators, but the need to talk to a human to solve most problems is still relevant, which means that there is still no ideal, fully satisfying all customer needs conversational artificial intelligence (AI). To improve the quality of service by the robot it is necessary to improve the quality of speech recognition of the interlocutor by the machine, for which neural networks are used. This article discusses the current state of development of neural networks, as well as artificial intelligence, which can be used to recognize human speech using a formant approach.