Views
No views yet
| Detail | Value |
|---|---|
| Model | VITS (Variational Inference with adversarial learning for end-to-end Text-to-Speech) |
| Language | Sinhala (සිංහල) |
| Epochs | 300 |
| Final mel loss | ~18.92 |
| Dataset | Multi-speaker TTS Dataset Sinhala |
| GPU | NVIDIA A100-80GB (via Modal) |
| Training time | ~3.2 hours |
| Framework | Coqui TTS 0.27.5 |
1from huggingface_hub import hf_hub_download
2from TTS.utils.synthesizer import Synthesizer
3
4config_path = hf_hub_download(repo_id="ngpsanjaya/vits-sinhala", filename="config.json")
5model_path = hf_hub_download(repo_id="ngpsanjaya/vits-sinhala", filename="model.pth")
6
7synthesizer = Synthesizer(
8 tts_checkpoint=model_path,
9 tts_config_path=config_path,
10 use_cuda=True,
11)
12
13wav = synthesizer.tts("ආයුබෝවන්")1import numpy as np
2import soundfile as sf
3
4sf.write("output.wav", np.array(wav), synthesizer.tts_config.audio.sample_rate)