In this paper, we propose a new adapter network for adapting a pre-train...
We propose U-Singer, the first multi-singer emotional singing voice
synt...
We propose OUR-GAN, the first one-shot ultra-high-resolution (UHR) image...
We propose a novel high-fidelity expressive speech synthesis model, UniT...
We propose an end-to-end speech synthesizer, Fast DCTTS, that synthesize...
We present a novel high-fidelity real-time neural vocoder called VocGAN....
In capsule networks, the routing algorithm connects capsules in consecut...
To address the issue of catastrophic forgetting in neural networks, we
p...