Mistral AI Releases Open Source Text-to-Speech Model

Paris-based Mistral AI has boldly entered the competitive voice AI arena by releasing a powerful new text-to-speech model—and giving it away for free. The company claims its model outperforms offerings from established leader ElevenLabs. By open-sourcing the model weights, Mistral is adopting a strategy that has fueled its success in the large language model space, applying pressure on competitors who typically charge for access to high-quality voice synthesis. This move directly challenges not just ElevenLabs, but also enterprise giants like IBM Watsonx and Google's Chirp. It intensifies competition in a sector crucial for applications in accessibility tools, media production, customer service, and interactive AI assistants. Mistral's play could accelerate innovation and drive down costs across the entire voice AI market.