SPEAKER IDENTIFICATION METHOD IN ARTIFICIAL INTELLIGENCE SECRETARY SERVICE FUSED WITH CONTEXT-DEPENDANT SPEAKER RECOGNITION AND CONTEXT-INDEPENDENT SPEAKER RECOGNITION AND VOICE RECOGNITION DEVICE USED THEREIN

摘要

Disclosed are a speaker identification method in artificial intelligence secretary service fused with context-dependent speaker identification and context-independent speaker identification and a speech recognition device used therein. The present invention is implemented through a process in which a speech recognition device stores users utterance voice information for a predetermined call word in a users directory and identifies the user based on the call word utterance voice information from the user thereby accumulatively storing utterance voice information of the followed unstructured natural language instruction in the users directory and generates a user voice parameter for context-independent speaker identification based on utterance voice information of an atypical natural language instruction cumulatively stored in the users directory. According to the present invention, it is possible to prevent a person who does not have a legitimate use right from being provided with an artificial intelligence secretary service using an artificial intelligent speaker. In addition, according to the present invention, it is possible to individually identify a plurality of users using the artificial intelligent speaker, thereby preventing a malfunction due to the inability to identify the user and providing a personalized service for each user.;COPYRIGHT KIPO 2019