Title :
Prosody Modeling and Eigen-Prosody Analysis for Robust Speaker Recognition
Author :
Chen, Zi-He ; Liao, Yuan-Fu ; Juang, Yau-Tarng
Author_Institution :
Dept. of Electr. Eng., Nat. Central Univ., Taoyuan, Taiwan
fDate :
March 18-23, 2005
Keywords :
eigenvalues and eigenfunctions; error statistics; mobile handsets; speaker recognition; speech processing; vector quantisation; HTIMIT database; VQ; a priori knowledge interpolation; eigen-prosody analysis; full text document retrieval; handset distortion compensator; limited training/test data; performance degradation; prosodic feature contours; prosody modeling; relative error rate reduction; robust speaker recognition; speaker identification; unseen handset mismatch; vector quantization; Cepstral analysis; Degradation; Hidden Markov models; Interpolation; Loudspeakers; Maximum likelihood linear regression; Robustness; Speaker recognition; Speech; Telephone sets;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on
Print_ISBN :
0-7803-8874-7
DOI :
10.1109/ICASSP.2005.1415081
