SinhalaVITS Playground
An Open Source Sinhala Text-to-Speech Model
by Dialog Axiata PLC in collaboration with
Dialog-University of Moratuwa Mobile Communications Research Laboratory
✍️ Input Text
🗣️ Choose Voice Model
Dropdown
🎧 Audio Result
Enter text and press 'Generate Speech' to begin
⚠️ Please Note:
- Do not use English letters or numerals in the input. Use only Sinhala Unicode text.
- Use proper punctuation (.,?) to help the model with phrasing.
- Keep input short for a single request to speed up synthesizing.
- * The speech quality depends on the trained dataset and may not sound completely natural.
This playground is developed by Kasun Ranasinghe (@KasunUoM)