Multi-modal Adversarial Training for Zero-Shot Voice Cloning
Abstract
A text-to-speech (TTS) model trained to reconstruct speech given text tends towards predictions that are close to the average characteristics of a dataset, failing to model the variations that make human speech sound natural. This problem is magnified for zero-shot voice cloning, a task that requires training data with high variance in speaking styles. We build off of recent works which have used Generative Advsarial Networks (GAN) by proposing a Transformer encoder-decoder architecture to conditionally discriminates between real and generated speech features. The discriminator is used in a training pipeline that improves both the acoustic and prosodic features of a TTS model. We introduce our novel adversarial training technique by applying it to a FastSpeech2 acoustic model and training on Libriheavy, a large multi-speaker dataset, for the task of zero-shot voice cloning. Our model achieves improvements over the baseline in terms of speech quality and speaker similarity. Audio examples from our system are available online.
Cite
@article{arxiv.2408.15916,
title = {Multi-modal Adversarial Training for Zero-Shot Voice Cloning},
author = {John Janiczek and Dading Chong and Dongyang Dai and Arlo Faria and Chao Wang and Tao Wang and Yuzong Liu},
journal= {arXiv preprint arXiv:2408.15916},
year = {2024}
}
Comments
Accepted at INTERSPEECH 2024