Abstract
Compared with conventional vocoders, deep neural network-based raw audio generative models, such as WaveNet and SampleRNN, can more naturally synthesi......
小提示：本篇文献需要登录阅读全文，点击跳转登录
Full Text Link
