VOICE RECOGNITION SERVER AND VOICE RECOGNITION METHOD

摘要

PURPOSE: A voice recognition server and a method thereof are provided to apply terminal information of a user terminal and voice information of a user and perform voice recognition through a relevant sound model, thereby providing a voice recognition result of a state where a channel environment of the user terminal is matched to a training environment of the sound model. CONSTITUTION: A sound model storage part(220) stores multiple sound models. A sound model extraction part(260) extracts a relevant sound model among the stored multiple sound models based on the gender of a user of a user terminal determined by a user gender determination part(240) and terminal information verified by a terminal information verifying part(250). A decoding part(270) applies the extracted sound model and recognizes the voice of the user. [Reference numerals] (210) Control part; (220) Sound model storage part; (230) Communication part; (240) User gender determination part; (250) Terminal information verifying part; (260) Sound model extraction part; (270) Decoding part; (AA) Terminal/gender sound model 1; (BB) Terminal/gender sound model 2; (CC) Terminal/gender sound model N