Varianceflow: High-Quality and Controllable Text-to-Speech using Variance Information via Normalizing Flow
Authors:
- 23 May 2022
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 7477-7481
Abstract
There are two types of methods for non-autoregressive text-to-speech models to learn the one-to-many relationship between text and speech effectively. The first one is to use an advanced generative framework such as normalizing flow (NF). The second one is to use variance information such as pitch or energy together when generating speech. For the second type, it is also possible to control the variance factors by adjusting the variance values provided to a model. In this paper, we propose a novel model called VarianceFlow combining the advantages of the two types. By modeling the variance with NF, VarianceFlow predicts the variance information more precisely with improved speech quality. Also, the objective function of NF makes the model use the variance information and the text in a disentangled manner resulting in more precise variance control. In experiments, VarianceFlow shows superior performance over other state-of-the-art TTS models both in terms of speech quality and controllability.Keywords
All Related Versions
This publication has 6 references indexed in Scilit:
- FastPitchFormant: Source-Filter Based Decomposed Modeling for Speech Synthesis2021
- GANSpeech: Adversarial Training for High-Fidelity Multi-Speaker Speech Synthesis2021
- Neural Speech Synthesis with Transformer NetworkProceedings of the AAAI Conference on Artificial Intelligence, 2019
- Introducing Parselmouth: A Python interface to PraatJournal of Phonetics, 2018
- Natural TTS Synthesis by Conditioning Wavenet on MEL Spectrogram Predictions2018
- Reducing F0 Frame Error of F0 tracking algorithms under noisy conditions with an unvoiced/voiced classification frontend2009