April 22, 2025:
Dia: Open Source TTS Model Challenges Tech Giants - Nari Labs has unveiled Dia, a 1.6 billion parameter open source text-to-speech model. Rivaling major players like ElevenLabs and OpenAI, Dia offers features like emotional tone, speaker tagging, and nonverbal audio cues, supporting nuanced and natural dialogue. The model is available for free on platforms like Hugging Face and GitHub under an Apache 2.0 license, encouraging commercial use.
With notable backing from Google's TPU Research Cloud, Dia positions itself as a formidable competitor in the TTS landscape, focusing on accessibility and expressive quality.
