Research of Language Recognition Methods Based on Machine Learning

摘要

This article is focused on speech recognition methods research and analysis of their efficiency. It presents a review of the most known ways to convert a person's speech into text, their effectiveness, and complexity in exploitation. The article scribes the application of machine learning in speech recognition tasks. A program was developed with a neural network for the research realization – a new technique of speech translation automation. The experiment provides information about comparing methods of speech recognition and analyzing the results of using them. Conclusions about the further realization of the research were drawn based on these results.