Low-resource expressive text-to-speech using data augmentation
This addresses the high data requirement for expressive TTS, benefiting applications needing personalized or stylized speech with minimal recording effort, though it is incremental as it builds on existing augmentation and fine-tuning techniques.
The paper tackles the problem of building expressive text-to-speech systems with limited target speaker recordings by proposing a 3-step data augmentation method using voice conversion, training, and fine-tuning, achieving significant improvements in perceived speech quality with as little as 15 minutes of recordings.
While recent neural text-to-speech (TTS) systems perform remarkably well, they typically require a substantial amount of recordings from the target speaker reading in the desired speaking style. In this work, we present a novel 3-step methodology to circumvent the costly operation of recording large amounts of target data in order to build expressive style voices with as little as 15 minutes of such recordings. First, we augment data via voice conversion by leveraging recordings in the desired speaking style from other speakers. Next, we use that synthetic data on top of the available recordings to train a TTS model. Finally, we fine-tune that model to further increase quality. Our evaluations show that the proposed changes bring significant improvements over non-augmented models across many perceived aspects of synthesised speech. We demonstrate the proposed approach on 2 styles (newscaster and conversational), on various speakers, and on both single and multi-speaker models, illustrating the robustness of our approach.