A Variational EM Algorithm for the Separation of Time-Varying Convolutive Audio Mixtures
This work addresses audio source separation for applications like speech enhancement, but it is incremental as it builds on existing probabilistic and NMF-based methods.
The paper tackled the problem of separating audio sources from time-varying convolutive mixtures by proposing a probabilistic framework with a variational EM algorithm, resulting in outperformance over a state-of-the-art baseline method in experiments on simulated data.
This paper addresses the problem of separating audio sources from time-varying convolutive mixtures. We propose a probabilistic framework based on the local complex-Gaussian model combined with non-negative matrix factorization. The time-varying mixing filters are modeled by a continuous temporal stochastic process. We present a variational expectation-maximization (VEM) algorithm that employs a Kalman smoother to estimate the time-varying mixing matrix, and that jointly estimate the source parameters. The sound sources are then separated by Wiener filters constructed with the estimators provided by the VEM algorithm. Extensive experiments on simulated data show that the proposed method outperforms a block-wise version of a state-of-the-art baseline method.