Just-in-Time (JiT) is a training-free, model-agnostic acceleration framework for Diffusion Transformers (DiTs). By leveraging a Spatially Approximated Generative ODE (SAG-ODE) and a Deterministic ...
The Effect of Training Dataset Size on Discriminative and Diffusion-Based Speech Enhancement Systems
Abstract: The performance of deep neural network-based speech enhancement systems typically increases with the training dataset size. However, studies that investigated the effect of training dataset ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results