EUROSPEECH 2003 - INTERSPEECH 2003
8th European Conference on Speech Communication and Technology

Geneva, Switzerland
September 1-4, 2003

        

Speaker Recognition Using MPEG-7 Descriptors

Hyoung-Gook Kim, Edgar Berdahl, Nicolas Moreau, Thomas Sikora

Technische Universitšt Berlin, Germany

Our purpose is to evaluate the efficiency of MPEG-7 audio descriptors for speaker recognition. The upcoming MPEG-7 standard provides audio feature descriptors, which are useful for many applications. One example application is a speaker recognition system, in which reduced-dimension log-spectral features based on MPEG-7 descriptors are used to train hidden Markov models for individual speakers. The feature extraction based on MPEG-7 descriptors consists of three main stages: Normalized Audio Spectrum Envelope (NASE), Principal Component Analysis (PCA) and Independent Component Analysis (ICA). An experimental study is presented where the speaker recognition rates are compared for different feature extraction methods. Using ICA, we achieved better results than NASE and PCA in a speaker recognition system.

Full Paper

Bibliographic reference.  Kim, Hyoung-Gook / Berdahl, Edgar / Moreau, Nicolas / Sikora, Thomas (2003): "Speaker recognition using MPEG-7 descriptors", In EUROSPEECH-2003, 489-492.