Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.
| Paper: | SP-L4.4 |
| Session: | Higher-Level Knowledge in Speaker Recognition |
| Time: | Wednesday, May 19, 16:30 - 16:50 |
| Presentation: |
Lecture |
| Topic: |
Speech Processing: Speaker Recognition |
| Title: |
APPLYING ARTICULATORY FEATURES TO TELEPHONE-BASED SPEAKER VERIFICATION |
| Authors: |
Ka-Yee Leung; Hong Kong Polytechnic University | | |
| | Man-Wai Mak; Hong Kong Polytechnic University | | |
| | Sun-Yuan Kung; Princeton University | | |
| Abstract: |
This paper presents an approach that uses articulatory features(AFs) derived from spectral features for telephone-basedspeaker verification. To minimize the acoustic mismatch causedby different handsets, handset-specific normalization isapplied to the spectral features before the AFs are extracted.Experimental results based on 150 speakers using 10 differenthandsets show that AFs contain useful speaker-specificinformation for speaker verification and the use ofhandset-specific normalization significantly lowers theerror rates under the handset mismatched conditions.Results also demonstrate that fusing the scores obtainedfrom an AF-based system with those obtained from a spectralfeature-based (MFCC) system helps lower the error rates ofthe individual systems. |
| |
| Back | |