SPEECH RECOGNITION DEVICE AND METHOD USING INDIVIDUAL SOUND MODEL

摘要

Disclosed are a voice recognition apparatus and a method using an individual acoustic model. According to one embodiment of the present invention, the voice recognition apparatus using an individual acoustic model includes: a collection unit which collects voice data; a learning unit which performs learning based on the voice data and generates an individual voice model of a speaker who is a main agent of the voice data; a storage unit which stores a common acoustic model, the voice data, and the individual acoustic model; and a voice recognition unit which selects the common acoustic model or individual acoustic model based on a voice data collection volume of the speaker in accordance with a voice recognition request of the speaker to perform voice recognition.