Nov, 2022
Stutter-TTS:受控制的口吃语音合成与改进的口吃语音识别
Stutter-TTS: Controlled Synthesis and Improved Recognition of Stuttered Speech
Xin Zhang, Iván Vallés-Pérez, Andreas Stolcke, Chengzhu Yu, Jasha Droppo...
TL;DRStutter-TTS is a neural model that can synthesize different types of stuttering utterances for improving automatic speech recognition on stuttered speech by introducing additional tokens to represent stuttering characteristics during training.