Thunder : Unified Regression-Diffusion Speech Enhancement with a Single Reverse Step using Brownian Bridge
This work addresses inference speed for speech enhancement users, but it is incremental as it builds on existing regression-diffusion methods.
The paper tackles the problem of slow inference time in diffusion-based speech enhancement by proposing Thunder, a unified regression-diffusion model using a Brownian bridge process, which achieves competitive performance with a more compact model size and fewer reverse steps.
Diffusion-based speech enhancement has shown promising results, but can suffer from a slower inference time. Initializing the diffusion process with the enhanced audio generated by a regression-based model can be used to reduce the computational steps required. However, these approaches often necessitate a regression model, further increasing the system's complexity. We propose Thunder, a unified regression-diffusion model that utilizes the Brownian bridge process which can allow the model to act in both modes. The regression mode can be accessed by setting the diffusion time step closed to 1. However, the standard score-based diffusion modeling does not perform well in this setup due to gradient instability. To mitigate this problem, we modify the diffusion model to predict the clean speech instead of the score function, achieving competitive performance with a more compact model size and fewer reverse steps.