Speech recognition and word conversion of speaker in congress

摘要

A conference hall speaker voice recognition and text information conversion system according to an embodiment of the present invention includes: a speaker information DB for generating and storing speaker information, which is information of speakers speaking at the conference hall; A feature information DB for generating and storing voice feature information and pronunciation feature information of each of the speakers; An identification module for identifying a speaker speaking in the conference hall; A character conversion module for generating and converting character conversion information for the speech of the speaker identified by the identification module; and an update module for updating pronunciation characteristic information of the speaker using the character conversion information generated by the character conversion module; It is characterized in that it includes. According to the present invention, the speaker can be identified based on the speaker's pronunciation characteristics, and the speaker's speech can be voice-recognized and converted into text information. In addition, according to the present invention, since voice recognition is performed using data for each speaker, the voice recognition rate can be increased.