SPEAKER IDENTIFICATION DEVICE, SPEAKER IDENTIFICATION METHOD, AND SPEAKER IDENTIFICATION PROGRAM

摘要

A speaker identification device comprises: a primary speaker identification part (1) which computes, for each pre-stored registered speaker, a score which denotes the similarity between inputted speech and speech of the registered speakers; a similar speaker selection part (2) which selects a plurality of the registered speakers as similar speakers according to the size of the scores thereof; a learning part (3) which creates a classifier for each similar speaker, with the speech of one similar speaker among the similar speakers being a positive instance and the speech of the other similar speakers being negative examples; and a secondary speaker identification part (4) which computes, for each classifier, a score of the classifier with respect to the inputted speech, and outputs an identification result.