摘要 |
A system for recognizing a sound according to an embodiment of the present invention comprises: a sound recognition service device for receiving a sound recognition request including sound data and a user key from a terminal, generating a word row corresponding to the sound data according to a sound model and a conversion parameter corresponding to the user key, and transmitting the word row to the terminal; and a user cluster device for storing the sound model and conversion parameter, providing the sound model and conversion parameter to the sound recognition service device, and renewing the sound model and conversion parameter according to the word row. |