Speaker recognition using multiple kernel learning based on conditional entropy minimization

We applied a multiple kernel learning (MKL) method based on information-theoretic optimization to speaker recognition. Most of the kernel methods applied to speaker recognition systems require a suitable kernel function and its parameters to be determined for a given data set. In contrast, MKL eliminates the need for strict determination of the kernel function and parameters by using a convex combination of element kernels. In the present paper, we describe an MKL algorithm based on conditional entropy minimization (MCEM). We experimentally veri ed the effectiveness of MCEM for speaker classi cation; this method reduced the speaker error rate as compared to conventional methods.

• Type: Journal
• Year: 2011
• Where: ICASSP
• Authors: Tetsuji Ogawa, Hideitsu Hino, Nima Reyhani, Noboru Murata, Tetsunori Kobayashi
• Keywords: ICASSP 2011, Kernel Function, Signal Processing, Speaker Classi Cation, Suitable Kernel Function