Ninth International Conference on Spoken Language Processing

Pittsburgh, PA, USA
September 17-21, 2006

Enhancing the Performance of a GMM-Based Speaker Identification System in a Multi-Microphone Setup

Andreas Stergiou, Aristodemos Pnevmatikakis, Lazaros C. Polymenakos

Athens Information Technology, Greece

In this paper the speaker identification system developed at Athens Information Technology is presented. It is based on the Gaussian Mixture modeling of the Mel-Frequency Cepstral Coefficients of speech. Starting from this basic algorithm, we describe and discuss two significant modifications that have resulted in performance enhancements, in terms of both processing speed and identification accuracy. We present the performance of our system in the recent CLEAR 2006 evaluation workshop and also discuss approaches to further improve our system by fusing decisions derived from a multitude of sensors in a multi-microphone setup.

Full Paper

Bibliographic reference.  Stergiou, Andreas / Pnevmatikakis, Aristodemos / Polymenakos, Lazaros C. (2006): "Enhancing the performance of a GMM-based speaker identification system in a multi-microphone setup", In INTERSPEECH-2006, paper 1608-Wed1A1O.3.