metadata
license: cc-by-nc-4.0
language:
- hak
pipeline_tag: text-to-speech
Model Card for taiwanese-hakka-tts-sixian-hailu-mix
A experimental model to try speaker encoder.
Model Details
- YourTTS
- sample rate: 16k
- Training data contain four persons
- Use speaker encoder output as speaker embedding
- Use language embedding
Model Sources
- Repository: coqui-ai/TTS
- Demo: Hugging Face Space