F5-TTS-Emotion-CFG-1
This repository provides pretrained F5-TTS emotion-conditioned checkpoints for emotion control.
The models enable emotional speech synthesis.
📦 Available Checkpoints
model_emo.pt – Emotion-conditioned modelmodel_0.pt – Neutral / baseline model
🎯 Task
- Audio / Text-to-Speech
- Emotion-controlled speech synthesis
🔗 Project Repository and Usage
📄 License
Attribution-NonCommercial 4.0 International
Mention: the code is released under MIT license. However, since the model weights have been trained on a non-comercial dataset (ESD), the checkpoints are also non-comerical.
