ClArTTS

ClArTTS是一个专为古典阿拉伯语文本到语音合成设计的大型数据集，由MBZUAI UAE创建。该数据集包含约12小时的单一男性演讲者录音，采样率为40100 kHz，共10,334条语音片段。数据来源于LibriVox有声书，经过精细处理、分割和手动转录标注。ClArTTS旨在支持阿拉伯语的端到端TTS系统开发，解决现有阿拉伯语TTS研究中数据集不足的问题。

ClArTTS is a large-scale dataset specifically designed for classical Arabic text-to-speech synthesis, created by MBZUAI UAE. It contains approximately 12 hours of recordings from a single male speaker, with a sampling rate of 40100 kHz and a total of 10,334 speech segments. The data is sourced from LibriVox audiobooks, and has undergone meticulous processing, segmentation and manual transcription annotation. ClArTTS aims to support the development of end-to-end TTS systems for Arabic, addressing the issue of insufficient datasets in existing Arabic TTS research.