AI Breakthrough: Model Screams in Terror

Nari Labs’ Dia-1.6B AI Model Challenges Industry Giants

• Dia-1.6B is an open-source text-to-speech model with 1.6 billion parameters.
• The model claims to outperform ElevenLabs and Sesame in emotional speech synthesis.
• It can produce realistic dialogue, including laughter, coughs, and screams.
• Dia-1.6B runs in real-time on a single GPU with 10GB of VRAM.
• The model is available under the Apache 2.0 license on Hugging Face and GitHub.

Nari Labs’ Dia-1.6B model, with its ability to generate emotionally expressive speech using just 1.6 billion parameters, claims superiority over established competitors like ElevenLabs and Sesame by synthesizing realistic nonverbal expressions such as laughter and screams while running efficiently on a single GPU.