INTERSPEECH 2011 12th Annual Conference of the International Speech Communication Association Florence, Italy August 27-31. 2011

Kernel Alignment Maximization for Speaker Recognition Based on High-Level Features

Szymon Drgas, Adam Dabrowski

Poznan University of Technology, Poland

In this paper text-independent automatic speaker verification based on support vector machines is considered. A generalized linear kernel training method based on kernel alignment maximization is proposed. First, kernel matrix decomposition into a sum of maximally aligned directions in the input space is performed and this decomposition is spectrally optimized. The method was evaluated for high-level speaker features: prosodic, articulatory and lexical. The experiments were undertaken employing Switchboard corpus. The proposed algorithm gave equal error rate (EER) reduction up to 23%.

Full Paper

Bibliographic reference.  Drgas, Szymon / Dabrowski, Adam (2011): "Kernel alignment maximization for speaker recognition based on high-level features", In INTERSPEECH-2011, 489-492.