INTERFACE WITH COMBINATION OF VOICE RECOGNITION TECHNIQUE AND MOTION/GESTURE RECOGNITION TECHNIQUE TO BE APPLIED TO ANY LANGUAGE CAPABILITIES INCLUDING GENDER, AGE, AND DISABILITY, AND USER INTENTION INPUT METHOD USING SAME

摘要

Provided is a voice-motion combination interface (100) for a device. The voice-motion combination interface (100) is used for delivering a user intention to the device, and includes a voice recognizing part (110), a motion sensing part (140), and an intention identifying part (150). The voice recognizing part (110) includes a sound model pool (120) and a voice recognition engine (130). The sound model pool (120) includes: two sound model sub-pools (122, 123) configured suitably for standard pronunciations; and three sound model sub-pools (124, 126, 128) specialized for nonstandard pronunciations. One among the sound model sub-pools (122, 123) and the sound model sub pools (124, 126, 128) is selectively applied. The motion sensing part (140) is able to be configured to recognize physical force applied by the user to the body of the device and a motion, a gesture, or behavior of the user, influencing various sensors embedded in the device. The motion sensing part (140) is able to be configured to recognize the user manipulating a component of the device. The intention identifying part (150) is configured to identify a specific intention to be inputted by the user by combining an intention of the user, determined by the motion sensing part (140), with a candidate group of the user intention extracted by the voice recognizing part (110).;COPYRIGHT KIPO 2018