Source Localization using a Spatial Kernel based Covariance Model and Supervised Complex Nonnegative Matrix Factorization
* Presenting author
This paper presents an algorithm for source localization using a beamforming-inspired spatial covariance model (SCM) and complex non-negative matrix factorization (CNMF). The spatial properties are modeled as the weighted sum of spatial kernels which encode the phase differences between microphones for every possible source location in a grid. The actual localization for each individual source in the multichannel mixture is estimated using complex-valued non-negative matrix factorization (CNMF) where each source spectrogram is modeled using a dictionary of spectral patterns learned a priori from training material. Localization performance of the proposed system is evaluated using a multi-channel dataset with configurations (number of simultaneous sources, reverberation time, microphones spacing, source types and spatial locations of the sources). Finally, a comparison to other state-of-the-art localization methods is performed, showing competitive localization performance.