1 paper
Chengxi Lei, Satwinder Singh, Feng Hou +2
Most of the current speech data augmentation methods operate on either the raw waveform or the amplitude spectrum of speech. In this paper, we propose a novel speech data augmentat…