Adversarial Speaker-Consistency Learning Using Untranscribed Speech Data for Zero-Shot Multi-Speaker Text-to-Speech

TL;DR

This paper introduces adversarial speaker-consistency learning (ASCL), a novel approach that improves zero-shot multi-speaker TTS by using untranscribed speech data and adversarial training to enhance speaker similarity and quality.

Contribution

The paper proposes a new adversarial learning framework that leverages untranscribed speech data to address speaker domain shift in zero-shot multi-speaker TTS.

Findings

Improved speaker similarity in zero-shot TTS

Enhanced speech quality over baseline models

Effective use of untranscribed data in training

Abstract

Several recently proposed text-to-speech (TTS) models achieved to generate the speech samples with the human-level quality in the single-speaker and multi-speaker TTS scenarios with a set of pre-defined speakers. However, synthesizing a new speaker's voice with a single reference audio, commonly known as zero-shot multi-speaker text-to-speech (ZSM-TTS), is still a very challenging task. The main challenge of ZSM-TTS is the speaker domain shift problem upon the speech generation of a new speaker. To mitigate this problem, we propose adversarial speaker-consistency learning (ASCL). The proposed method first generates an additional speech of a query speaker using the external untranscribed datasets at each training iteration. Then, the model learns to consistently generate the speech sample of the same speaker as the corresponding speaker embedding vector by employing an adversarial learning scheme. The experimental results show that the proposed method is effective compared to the baseline in terms of the quality and speaker similarity in ZSM-TTS.