Àá½Ã¸¸ ±â´Ù·Á ÁÖ¼¼¿ä. ·ÎµùÁßÀÔ´Ï´Ù.
KMID : 0917520000070030185
Journal of Speech Sciences
2000 Volume.7 No. 3 p.185 ~ p.195
Parameters Comparison in the speaker Indentification under the Noisy Environments


Abstract
This paper seeks to compare the feature parameters used in speaker identification systems under noisy environments. The feature parameters compared are LP cepstrum (LPCC), Cepstral mean subtraction(CMS), Pole-filtered CMS(PFCMS), Adaptive compo-nent weighted cepstrum(ACW) and Postfilter cepstrum(PF). The GMM-based text independent show that the LPCC parameter is adequate for modelling the speaker in the matched environments between train and test stages.But in the mismatched training and testing condition,s modified parameters are preferable the LPCC. Especially CMS and PFCMS parameters are more effective for the microphone mismatching conditions while ACW and PF parameters are good for more noisy mismatches.
Keywords : speaker indentification, GMM, LPCC, CMS, PFCMS, ACW, PF
KEYWORD
FullTexts / Linksout information
Listed journal information