ON-LINE SPEAKER RECOGNITION METHOD AND APPARATUS FOR THEREOF

摘要

An online speaker recognition method and an apparatus therefor are provided to enable a speaker model adaptation module to update continuously a speaker model stored in a memory unit by using the voice data of a speaker inputted continuously by contents, thereby being adapted to the voice data of the speaker which is registered according to a change of time and environment. An online speaker recognition method comprises the following steps of: receiving basic data of a speaker and the voice data of the speaker by using contents which requests the continuous voice response of the speaker(S100,S105); extracting only the voice of the speaker from the voice data(S115); extracting a specific vector necessary for the speaker recognition from the speaker's voice(S120); creating a speaker model from the extracted specific vector(S125); and recognizing the speaker stored in the speaker model on the basis of information recognized from the input voice(S130).