System

Abstract

An object of a system according to an embodiment is to recognize speech data in which dialects and emotions are mixed with high accuracy and generate an optimal response.SOLUTION: A system according to an embodiment includes a feature point extraction unit, a distribution unit, a voice recognition unit, and a voice bot unit. The feature point extraction unit extracts feature points of dialects and emotions from the speech data. The distribution unit distributes the voice data to an appropriate voice recognition engine on the basis of the feature point extracted by the feature point extraction unit. The voice recognition unit analyzes the voice data distributed by the distribution unit. The voice bot unit generates an optimal response based on the text data analyzed by the voice recognition unit.SELECTED DRAWING: None