Kurdish end-to-end speech synthesis using deep neural networks
Authors
Sabat Salih Muhamad , Hadi Veisi , Aso Mahmudi , Abdulhady Abas Abdullah , Farhad Rahimi
Natural Language Processing Journal
Vol. 8
pp. 100096
January 1, 2024
Abstract
A Central Kurdish end-to-end speech synthesis study comparing Tacotron 2 and WaveGlow configurations, including transfer learning from English and training from scratch. The work uses a single-speaker female Kurdish speech corpus associated with the AsoSoft corpus family.
Keywords
Central Kurdish
Sorani
Tacotron 2
WaveGlow
Speech Synthesis
Related Datasets
AsoSoft Speech Corpus for Central-Kurdish Text-To-Speech
2022
Public subset: approximately 1h02m and 522 text/audio pairs
WAV, TXT, RAR
A Central Kurdish single-speaker female speech corpus associated with AsoSoft. The full corpus is described as 21 hours and 10,979 sentences, but the public repository offers only an approximately …
Citation
Muhamad, S. S., Veisi, H., Mahmudi, A., Abdullah, A. A., & Rahimi, F. (2024). Kurdish end-to-end speech synthesis using deep neural networks. Natural Language Processing Journal, 8, 100096. https://doi.org/10.1016/j.nlp.2024.100096