fairseq S^2: A Scalable and Integrable Speech Synthesis Toolkit
This toolkit addresses the need for scalable and integrable speech synthesis tools for researchers and developers, but it is incremental as it builds upon the existing fairseq framework.
The paper introduces fairseq S^2, a toolkit for speech synthesis that implements autoregressive and non-autoregressive text-to-speech models with multi-speaker variants, and includes preprocessing tools to reduce data curation needs and automatic metrics for faster development.
This paper presents fairseq S^2, a fairseq extension for speech synthesis. We implement a number of autoregressive (AR) and non-AR text-to-speech models, and their multi-speaker variants. To enable training speech synthesis models with less curated data, a number of preprocessing tools are built and their importance is shown empirically. To facilitate faster iteration of development and analysis, a suite of automatic metrics is included. Apart from the features added specifically for this extension, fairseq S^2 also benefits from the scalability offered by fairseq and can be easily integrated with other state-of-the-art systems provided in this framework. The code, documentation, and pre-trained models are available at https://github.com/pytorch/fairseq/tree/master/examples/speech_synthesis.