Fecha de publicación: 20/08/2020 Fuente: Wipo "IOT" The present invention relates to a hybrid home speech recognition system and a method for providing the same. The method comprises: a first step of storing speech recognition information received from a microphone (210) of a sub IoT device (200)(I1) in a storage unit (330) as second speech recognition waveform; a second step of storing the speech recognition information in the storage unit (330) as (3-1)^th to (3-n)^th speech recognition waveform, wherein n is a natural number greater than or equal to 2; a third step of storing speech recognition information received in a microphone (210) of one or more other sub IoT devices (200) in the storage unit (330) as (4-1)^th to (4-m)^th speech recognition waveform, wherein m is a natural number greater than or equal to 2 and equal to or different from n; a fourth step of recognizing a character included in a control command; and a fifth step of transmitting the control command to a home IoT network (500). COPYRIGHT KIPO 2021