Kernel Machines Beat Deep Neural Networks on Mask-based Single-channel Speech Enhancement
This provides a more efficient alternative to deep neural networks for speech enhancement tasks, though it appears incremental as it applies existing kernel methods to a specific domain.
The authors tackled single-channel speech enhancement by applying a fast kernel method with automatic hyperparameter selection, achieving better performance than state-of-the-art deep neural networks on HINT and TIMIT datasets while requiring less training time.
We apply a fast kernel method for mask-based single-channel speech enhancement. Specifically, our method solves a kernel regression problem associated to a non-smooth kernel function (exponential power kernel) with a highly efficient iterative method (EigenPro). Due to the simplicity of this method, its hyper-parameters such as kernel bandwidth can be automatically and efficiently selected using line search with subsamples of training data. We observe an empirical correlation between the regression loss (mean square error) and regular metrics for speech enhancement. This observation justifies our training target and motivates us to achieve lower regression loss by training separate kernel model per frequency subband. We compare our method with the state-of-the-art deep neural networks on mask-based HINT and TIMIT. Experimental results show that our kernel method consistently outperforms deep neural networks while requiring less training time.