StoRM: A Diffusion-based Stochastic Regeneration Model for Speech Enhancement and Dereverberation

Understand

Diffusion models have shown a great ability at bridging the performance gap between predictive and generative approaches for speech enhancement.

• We have shown that they may even outperform their predictive counterparts for non-additive corruption types or when they are evaluated on mismatched conditions.
• However, diffusion models suffer from a high computational burden, mainly as they require to run a neural network for each reverse diffusion step, whereas predictive approaches only require one pass.
• As diffusion models are generative approaches they may also produce vocalizing and breathing artifacts in adverse conditions.