Abstract: The goal of this paper is to generate realistic audio with a lightweight and fast diffusion-based vocoder, named FreGrad. Our framework consists of the following three key components: (1) We ...

Abstract: The traditional vocoders have the advantages of high synthesis efficiency, strong interpretability, and speech editability, while the neural vocoders have the advantage of high synthesis ...

When it comes to the machines used to make their music, there are few who compare to Kraftwerk in the secrecy stakes. The iconic German synth masters' almost complete lack of studio interviews on the ...