metadata
license: mit
datasets:
- ryota-komatsu/libritts-r-mhubert-2000units
language:
- en
base_model:
- ryota-komatsu/fastspeech2_conformer_hifigan
Conditional Flow Matching-based acoustic model with a HiFi-GAN vocoder.
This is a model repository of a GitHub project.
The model was trained on 16 kHz downsampled LibriTTS-R and EXPRESSO HuBERT units.