Speaker Recognition Using Principal Component Analysis
This paper proposes a new feature vector-Mel Frequency Principal Coefflctent(MFPC), applied to speaker recognition. It is derived by performing Principal Component Analysis on the Mel Scale Spectrum Vector. Compared with conventional Mel Frequency Cepstrum Coefficient, MFPC efficiently exploited the correlation information among different frequency channels. These correlations, which is mainly caused by the vocal tract resonance, have been found to vary consistently from one speaker to another. And we select these feature coefficients according to their Fisher Ratio, which will guarantee the largest discriminability between classes in the given dimensionality. Finally, we implement a textindependent speaker recognition system. It uses Vector Quantization to design codebooks of given reference speakers. The experiment results demonstrate that our proposed feature vector has characteristics of compactness, large discriminability and low redundancy.
Mel Frequency Principal Coefficient (MFPC) Principle Component Analysis (PCA) Vector Quantization (VQ) Fisher Ratio
Peilv Ding Liming Zhang
Dept.of Electronic Engineering Fudan University China, Shanghai, 200433
国际会议
8th International Conference on Neural Information Processing(ICONIP 2001)(第八届国际神经信息处理大会)
上海
英文
1430-1435
2001-11-14（万方平台首次上网日期，不代表论文的发表时间）