RAD-TTS++ Ukrainian
Overview
Type your text in Ukrainian and select a voice to synthesize speech using the RAD-TTS++ model and Vocos with 44100 Hz.
Usage
Here you will see how fast the model and the speaker is.
You can change the voice, speed, and other parameters.
Choose an example
| Text | Voice |
|---|
Authors
Follow them on social networks and contact if you need any help or have any questions:
| Yehor Smoliakov |
|---|
| https://t.me/smlkw in Telegram |
| https://x.com/yehor_smoliakov at X |
| https://github.com/egorsmkv at GitHub |
| https://huggingface.co/Yehor at Hugging Face |
| or use egorsmkv@gmail.com |
Gradio app uses:
Environment
- Python: 3.10.17 (main, Apr 9 2025, 18:21:32) [GCC 12.2.0]
- Torch device: cpu
Models
Acoustic model (Text-to-MEL)
- Name: RAD-TTS++ (DAP)
- Parameters: 229,103,605
- File size: 929.73 MB
Vocoder (MEL-to-WAVE)
- Name: Vocos
- Parameters: 13,985,282
- File size: 56.32 MB
Libraries
- vocos: 0.1.0
- gradio: 5.20.0
- huggingface_hub: 0.30.2
- spaces: 0.34.2
- torch: 2.6.0
- torchaudio: 2.6.0
- scipy: 1.15.2
- numba: 0.61.2
- librosa: 0.11.0