Learning to Maximize Speech Quality Directly Using MOS Prediction for Neural Text-to-Speech. (arXiv:2011.01174v4 [eess.AS] UPDATED)

Although recent neural text-to-speech (TTS) systems have achieved
high-quality speech synthesis, there are cases where a TTS system generates
low-quality speech, mainly caused by limited training data or information loss
during knowledge distillation. Therefore, we propose a novel method to improve
speech quality by training a TTS model under the supervision of perceptual
loss, which measures the distance between the maximum possible speech quality
score and the predicted one. We first pre-train a mean opinion score (MOS)
prediction model and …

arxiv learning mos prediction speech text