Development of speech recognition service Bambusa

Abstract

A test version of the Bambusa service was developed by the staff of the interdepartmental laboratory for machine learning and data mining. This service allows the user to process large volumes of audio and video information without losing the quality of perception. The Bambusa service is designed to transcribe speech into text and generate a short retelling of the received text. The features of the microservices architecture of the application are described, the characteristics of similar commercial services are given