IndicF5 Tamil Fine-Tuned Model
This is a fine-tuned version of ai4bharat/IndicF5 optimized for high-quality, natural Tamil speech synthesis and zero-shot voice cloning.
Model Details
- Base Model:
ai4bharat/IndicF5 - Fine-Tuning Dataset:
TamilVoiceCorpus(PureVox split) - Fine-Tuning Method: Adapter-style training (Transformer backbone frozen, text embedding and projection layers updated)
- Language: Tamil (ta)
Files Included
model_last.pt: Fine-tuned PyTorch checkpoint.vocab.txt: Character vocabulary mapping generated from the Tamil dataset.
- Downloads last month
- 27