Advances in Speech Vocoding for Text-to-Speech with Continuous Parameters

TL;DR

This paper introduces a flexible, continuous vocoder for TTS that reduces noise perception and employs neural sequence models like LSTM and GRU to enhance speech naturalness, achieving state-of-the-art results.

Contribution

It proposes a novel continuous vocoder with phase distortion-based noise masking and integrates neural sequence models for improved speech synthesis.

Findings

Achieved state-of-the-art speech synthesis performance.

Reduced perceptual noise impact in vocoded speech.

Enhanced naturalness with neural sequence modeling.

Abstract

Vocoders received renewed attention as main components in statistical parametric text-to-speech (TTS) synthesis and speech transformation systems. Even though there are vocoding techniques give almost accepted synthesized speech, their high computational complexity and irregular structures are still considered challenging concerns, which yield a variety of voice quality degradation. Therefore, this paper presents new techniques in a continuous vocoder, that is all features are continuous and presents a flexible speech synthesis system. First, a new continuous noise masking based on the phase distortion is proposed to eliminate the perceptual impact of the residual noise and letting an accurate reconstruction of noise characteristics. Second, we addressed the need of neural sequence to sequence modeling approach for the task of TTS based on recurrent networks. Bidirectional long short-term memory (LSTM) and gated recurrent unit (GRU) are studied and applied to model continuous parameters for more natural-sounding like a human. The evaluation results proved that the proposed model achieves the state-of-the-art performance of the speech synthesis compared with the other traditional methods.