Aurora 1.6b

🚀 Aurora-1.6B: Multilingual Emotion and Singing TTS Model

A fine - tuned version of Dia - 1.6B, trained on multilingual and singing datasets, enabling full emotion control and zero - shot voice cloning.

🚀 Quick Start

Aurora-1.6B is a powerful multilingual text - to - speech model. It's a fine - tuned version of Dia - 1.6B, offering features like multilingual support, emotion control, singing capabilities, zero - shot voice cloning, and nonverbal vocalizations.

✨ Features

• Multilingual Support
  It can generate natural speech in Italian, English, Polish, German, French, and more.
• Emotion Control
  You can use speaker tags or emotion tokens (e.g., [S1], [happy], [sad]) to adjust the expressiveness of the speech.
• Singing Capabilities
  By providing singing prompts or style references, it can generate melodic vocals.
• Zero - Shot Voice Cloning
  It can clone any speaker's voice from a short audio sample.
• Nonverbal Vocalizations
  You can embed realistic effects like (laughs), (coughs), or (sighs) directly in the text.

💻 Usage Examples

Basic Usage

from dia.model import Dia
import soundfile as sf

# Load the Aurora-1.6B model
model = Dia.from_pretrained("Lorenzob/aurora-1.6b")

# Generate a happy spoken line followed by singing
text = "[S1][happy] Hello world! Now sing 'Happy Birthday to You'"
audio = model.generate(text)

# Save output at 44.1 kHz
sf.write("output.wav", audio, 44100)

📄 License

This project is licensed under the Apache - 2.0 license.