DINO-VITS: Data-Efficient Zero-Shot TTS with Self-Supervised Speaker Verification Loss for Noise Robustness

TL;DR

DINO-VITS introduces a noise-robust zero-shot TTS system using self-supervised speaker verification loss, significantly improving voice cloning quality in noisy environments through dual-objective training.

Contribution

The paper proposes a novel dual-objective training method for zero-shot TTS that enhances noise robustness and speaker discriminability using self-supervised DINO loss.

Findings

Significant improvements in subjective quality under noisy conditions

Effective training on noisy, unlabeled data for better voice cloning

Outperforms traditional speaker-encoder-based TTS systems

Abstract

We address zero-shot TTS systems' noise-robustness problem by proposing a dual-objective training for the speaker encoder using self-supervised DINO loss. This approach enhances the speaker encoder with the speech synthesis objective, capturing a wider range of speech characteristics beneficial for voice cloning. At the same time, the DINO objective improves speaker representation learning, ensuring robustness to noise and speaker discriminability. Experiments demonstrate significant improvements in subjective metrics under both clean and noisy conditions, outperforming traditional speaker-encoderbased TTS systems. Additionally, we explore training zeroshot TTS on noisy, unlabeled data. Our two-stage training strategy, leveraging self-supervised speech models to distinguish between noisy and clean speech, shows notable advances in similarity and naturalness, especially with noisy training datasets, compared to the ASR-transcription-based approach.