论文部分内容阅读
A novel text independent speaker identification system is proposed. In the proposed system, the 12-order perceptual linear predictive cepstrum and their delta coefficients in the span of five frames are extracted from the segmented speech based on the met