Independent Low-Rank Matrix Analysis Based on Complex Student's $t$-Distribution for Blind Audio Source Separation
This work addresses blind source separation for audio processing, offering incremental improvements in separation quality and stability.
The paper tackled blind audio source separation by generalizing the source generative model in independent low-rank matrix analysis (ILRMA) to use an isotropic complex Student's t-distribution, which improved performance and stability over the conventional Gaussian-based method in experiments on music and speech tasks.
In this paper, we generalize a source generative model in a state-of-the-art blind source separation (BSS), independent low-rank matrix analysis (ILRMA). ILRMA is a unified method of frequency-domain independent component analysis and nonnegative matrix factorization and can provide better performance for audio BSS tasks. To further improve the performance and stability of the separation, we introduce an isotropic complex Student's $t$-distribution as a source generative model, which includes the isotropic complex Gaussian distribution used in conventional ILRMA. Experiments are conducted using both music and speech BSS tasks, and the results show the validity of the proposed method.