hierspeech++.json

{"model_name": "HierSpeech++", "display_name": "[HierSpeech++ ](https://sh-lee-prml.github.io/HierSpeechpp-demo/) [\ud83e\udd17](https://huggingface.co/spaces/LeeSangHoon/HierSpeech_TTS)[\ud83d\udcbb](https://github.com/sh-lee-prml/HierSpeechpp)[\ud83d\udcc4](https://arxiv.org/abs/2311.12454)", "model_tags": ["Normalizing Flow", "Reference-based (Speaker)", "Prosodic Correlates", "Adversarial", "Pretrained Text Encoder", "Non-autoregressive"], "web_url": "https://sh-lee-prml.github.io/HierSpeechpp-demo/", "hf_url": "https://huggingface.co/spaces/LeeSangHoon/HierSpeech_TTS", "code_url": "https://github.com/sh-lee-prml/HierSpeechpp", "paper_url": "https://arxiv.org/abs/2311.12454", "inference_details": "Default settings from repo: noise scale 0.33 for both, denoise ratio 0", "status": "pending"}