Parallel Tacotron 2: A Non-Autoregressive Neural TTS Model with Differentiable Duration Modeling
arXiv:2103.14574
Abstract
This paper introduces Parallel Tacotron 2, a non-autoregressive neural text-to-speech model with a fully differentiable duration model which does not require supervised duration signals. The duration model is based on a novel attention mechanism and an iterative reconstruction loss based on Soft Dynamic Time Warping, this model can learn token-frame alignments as well as token durations automatically. Experimental results show that Parallel Tacotron 2 outperforms baselines in subjective naturalness in several diverse multi speaker evaluations. Its duration control capability is also demonstrated.
arXiv commentSubmitted to INTERSPEECH 2021
References in corpus (10)
- Searching for Activation Functions
- FastSpeech: Fast, Robust and Controllable Text to Speech
- Pay Less Attention with Lightweight and Dynamic Convolutions
- Soft-DTW: a Differentiable Loss Function for Time-Series
- Non-Attentive Tacotron: Robust and Controllable Neural TTS Synthesis Including Unsupervised Duration Modeling
- End-to-End Adversarial Text-to-Speech
- Parallel Tacotron: Non-Autoregressive and Controllable TTS
- JDI-T: Jointly trained Duration Informed Transformer for Text-To-Speech without Explicit Alignment
- Fully-hierarchical fine-grained prosody modeling for interpretable speech synthesis
- EfficientTTS: An Efficient and High-Quality Text-to-Speech Architecture