Study on Closed-set Speaker Identification Based on Biomimetic Pattern Recognition

— In this PaPer.a new classifier of speaker identification has been proposed, which is based on Biomimetic pattern recognition (BPR). Distinguished from traditional speaker rec0gnition m ethods, such as DW T.HM M ,G M M .SV M and so on,the proposed clas— sifier is constructed by som e finite sub-space which is reasonable covering of the points in high dim ensional space according to distributing characteristic of speech feature points. It has been used in the system of speaker identification.Experiment results show that better efI_ect could be obtained especially w ith lesser sam ples. Furtherm ore。 the proposed clas sifier em ploys a m uch sim pler m odeling structure as com pared to the G M M .In addition.the basic idea“cognition”of Biomimetic pattern recognition(BPR) results in no requirem ent of retraining the old system for enrolling new speakers.