We present a Classical Arabic Text-to-Speech (ClArTTS) corpus to facilitate the development of end-to-end TTS systems for the Arabic language. The speech is extracted from a LibriVox audiobook, which is then processed, segmented, and manually transcribed and annotated. The ClArTTS corpus contains about 12 hours of speech from a single male speaker sampled at 40100 Hz. In this paper, we describe the process of corpus creation, details of corpus statistics, and a comparison with existing resources. Furthermore, we develop two TTS systems based on Grad-TTS and Glow-TTS and illustrate the performance of the resulting systems via subjective and objective evaluations. The ClArTTS corpus is publicly available at www.clartts.com for research purposes, along with the baseline TTS systems and an interactive demo.
Cite as: Kulkarni, A., Kulkarni, A., Shatnawi, S.A.M., Aldarmaki, H. (2023) ClArTTS: An Open-Source Classical Arabic Text-to-Speech Corpus. Proc. INTERSPEECH 2023, 5511-5515, doi: 10.21437/Interspeech.2023-2224
@inproceedings{kulkarni23_interspeech,
author={Ajinkya Kulkarni and Atharva Kulkarni and Sara Abedalmon'em Mohammad Shatnawi and Hanan Aldarmaki},
title={{ClArTTS: An Open-Source Classical Arabic Text-to-Speech Corpus}},
year=2023,
booktitle={Proc. INTERSPEECH 2023},
pages={5511--5515},
doi={10.21437/Interspeech.2023-2224}
}