A Cyclical Post-filtering Approach to Mismatch Refinement of Neural Vocoder for Text-to-speech Systems
This work addresses data mismatch issues for developers of text-to-speech systems, offering an incremental improvement to enhance existing systems economically.
The paper tackles mismatches in neural vocoders for text-to-speech systems by using a cyclic voice conversion model to generate matched training and testing data, resulting in improved speech quality as confirmed by objective and subjective experiments.
Recently, the effectiveness of text-to-speech (TTS) systems combined with neural vocoders to generate high-fidelity speech has been shown. However, collecting the required training data and building these advanced systems from scratch are time and resource consuming. An economical approach is to develop a neural vocoder to enhance the speech generated by existing or low-cost TTS systems. Nonetheless, this approach usually suffers from two issues: 1) temporal mismatches between TTS and natural waveforms and 2) acoustic mismatches between training and testing data. To address these issues, we adopt a cyclic voice conversion (VC) model to generate temporally matched pseudo-VC data for training and acoustically matched enhanced data for testing the neural vocoders. Because of the generality, this framework can be applied to arbitrary TTS systems and neural vocoders. In this paper, we apply the proposed method with a state-of-the-art WaveNet vocoder for two different basic TTS systems, and both objective and subjective experimental results confirm the effectiveness of the proposed framework.