Orpheus Exl2 4bit

🚀 Orpheus 3B 0.1 Finetuned

On March 18, 2025, we released our 3B Orpheus TTS model with additional finetunes. The code is available on GitHub: CanopyAI/Orpheus-TTS.

Orpheus TTS is a state-of-the-art, Llama-based Speech-LLM crafted for high-quality, empathetic text-to-speech generation. This model has undergone finetuning to achieve human-level speech synthesis, boasting exceptional clarity, expressiveness, and real-time streaming performance.

✨ Features

Model Capabilities

• Human-Like Speech: It offers natural intonation, emotion, and rhythm, outperforming SOTA closed-source models.
• Zero-Shot Voice Cloning: Clone voices without prior fine-tuning.
• Guided Emotion and Intonation: Control speech and emotion characteristics using simple tags.
• Low Latency: With approximately 200ms streaming latency for real-time applications, it can be reduced to around 100ms with input streaming.

Model Sources

• GitHub Repo: https://github.com/canopyai/Orpheus-TTS
• Blog Post: https://canopylabs.ai/model-releases
• Colab Inference Notebook: notebook link

🚀 Quick Start

Check out our Colab (link to Colab) or GitHub (link to GitHub) for instructions on how to perform easy inference on our finetuned models.

📄 License

This project is licensed under the Apache-2.0 license.

⚠️ Important Note

Do not use our models for impersonation without consent, misinformation or deception (including fake news or fraudulent calls), or any illegal or harmful activity. By using this model, you agree to follow all applicable laws and ethical guidelines. We disclaim responsibility for any use.