Speech enhancement with mixture-of-deep-experts with clean clustering pre-training
This work addresses speech enhancement for noisy environments, offering incremental improvements in robustness and efficiency.
The authors tackled single-microphone speech enhancement by proposing a mixture of deep experts (MoDE) architecture, which improved robustness to unfamiliar noise types and reduced test-time complexity.
In this study we present a mixture of deep experts (MoDE) neural-network architecture for single microphone speech enhancement. Our architecture comprises a set of deep neural networks (DNNs), each of which is an 'expert' in a different speech spectral pattern such as phoneme. A gating DNN is responsible for the latent variables which are the weights assigned to each expert's output given a speech segment. The experts estimate a mask from the noisy input and the final mask is then obtained as a weighted average of the experts' estimates, with the weights determined by the gating DNN. A soft spectral attenuation, based on the estimated mask, is then applied to enhance the noisy speech signal. As a byproduct, we gain reduction at the complexity in test time. We show that the experts specialization allows better robustness to unfamiliar noise types.