SpeechT5 Neural Text-to-Speech (TTS) Model
This model is a neural text-to-speech architecture based on Microsoft SpeechT5, fine-tuned for high-fidelity voice generation as part of the Hugging Face Audio Transformers Course (Unit 6).
🚀 Model Details
- Task: Text-to-Speech (Neural Vocoding)
- Architecture: SpeechT5 with character-level SentencePiece tokenization
- Status: Officially Verified & Certified
- Downloads last month
- 46
Evaluation results
- Lossself-reported0.150