Targeting at both high efficiency and performance, we propose AlignTTS to predict the mel-spectrum in parallel. AlignTTS is based on a Feed-Forward Transformer which generates mel-spectrum from a sequence of characters, and the duration of each character is determined by a duration predictor.Instead of adopting the attention mechanism in Transformer TTS to align text to mel-spectrum, the alignment loss is presented to consider all possible alignments in training by use of dynamic programming. Experiments on the LJSpeech dataset show that our model achieves not only state-of-the-art performance which outperforms Transformer TTS by 0.03 in mean option score (MOS), but also a high efficiency which is more than 50 times faster than real-time.
@article{arxiv.2003.01950,
title = {AlignTTS: Efficient Feed-Forward Text-to-Speech System without Explicit Alignment},
author = {Zhen Zeng and Jianzong Wang and Ning Cheng and Tian Xia and Jing Xiao},
journal= {arXiv preprint arXiv:2003.01950},
year = {2020}
}