Can this be refitted for noise reduction / audio restoration purposes? On hf they mention this is a "latent diffusion" model, I guess it implies there is a comp

Can this be refitted for noise reduction / audio restoration purposes? On hf they mention this is a "latent diffusion" model, I guess it implies there is a component inside that tries to recover noisy signal. [edit] If not, what free and open ML tools that can be used for restoration / deconvolution / denoising are there?