Abstract
A novel eigen-prosody analysis approach is proposed for robust speaker recognition under a mismatch handset environment. The idea is to convert the prosodie contours of a speaker's speech into sequences of prosody symbols, and transform the speaker recognition problem into a full-text document retrieval-similar task. Experimental results on the HTIMIT corpus have shown that, even though only few training/test data are available, about 32.2% relative error rate reduction could be achieved compared with the conventional Gaussian mixture model/cepstral mean subtraction approach.
Original language | English |
---|---|
Pages (from-to) | 1233-1235 |
Number of pages | 3 |
Journal | Electronics Letters |
Volume | 40 |
Issue number | 19 |
DOIs | |
State | Published - 16 Sep 2004 |