HiFiTTS-2: A Large-Scale High Bandwidth Speech Dataset

TL;DR

HiFiTTS-2 is a comprehensive high-bandwidth speech dataset derived from LibriVox audiobooks, enabling advanced TTS model training with detailed metadata and a robust data processing pipeline.

Contribution

The paper introduces HiFiTTS-2, a large-scale high-bandwidth speech dataset with a novel data processing pipeline and detailed metadata for improved TTS research.

Findings

Dataset supports high-quality zero-shot TTS models

Data pipeline effectively estimates bandwidth and segments speech

Enables flexible data filtering for various applications

Abstract

This paper introduces HiFiTTS-2, a large-scale speech dataset designed for high-bandwidth speech synthesis. The dataset is derived from LibriVox audiobooks, and contains approximately 36.7k hours of English speech for 22.05 kHz training, and 31.7k hours for 44.1 kHz training. We present our data processing pipeline, including bandwidth estimation, segmentation, text preprocessing, and multi-speaker detection. The dataset is accompanied by detailed utterance and audiobook metadata generated by our pipeline, enabling researchers to apply data quality filters to adapt the dataset to various use cases. Experimental results demonstrate that our data pipeline and resulting dataset can facilitate the training of high-quality, zero-shot text-to-speech (TTS) models at high bandwidths.