INTERSPEECH 2004 - ICSLP
This paper proposes a noise-robust speaker verification method augmented by fundamental frequency (F0). The paper first describes a noise-robust F0 extraction method using the Hough transform. Then, it proposes a robust speaker verification method using multi-stream HMMs which fuse the extracted F0 and cepstral features. Experiments are conducted using four-connected-digit utterances of Japanese by 37 male speakers recorded at five sessions over a half year period. The utterances are contaminated with white noise at various SNR levels. Experimental results show that the F0 features improve the verification performance in all SNR conditions.
Bibliographic reference. Iwano, Koji / Asami, Taichi / Furui, Sadaoki (2004): "Noise-robust speaker verification using F0 features", In INTERSPEECH-2004, 1417-1420.