With an innovative text-to-speech feature, low latency, and its own graphics and vector engine, Brilliant Labs aims to give developers more options for their Frame apps.
Startup Brilliant Labs has announced a number of new features and powerful tools for its Frame AI headset. Key updates include advanced text-to-speech capabilities, reduced latency for streaming responses, and the introduction of an in-house graphics and vector engine.
Developers can now offer a more intuitive interaction between Frame’s AI assistant Noa and users, ultimately creating more creative and innovative AI applications.
Improvements include:
Multimodal AI headset Frame takes off
Brilliant Labs recently unveiled the world’s first smart glasses with a built-in multimodal AI assistant. The 39-gram Frame comes with an always-on AI assistant called Noa.
The multimodal generative AI system can run GPT4, Stability AI, and Whisper AI models simultaneously. By analyzing interactions between the user and the AI assistant, Noa is designed to evolve and develop a unique personality.
With Frame, Brilliant Labs aims to take the use of generative AI from the laptop to the real world. Instead of typing in questions, the system uses input from the environment. Noa processes what its camera picks up, recognizes speech and sounds, translates in real time, or summarizes entire pages of a book.
It can also compare prices in stores by scanning clothing or other items. Frame is not capable of sophisticated mixed reality features like Apple Vision Pro or Meta Quest 3. The MicroOLEDs built into the frame display images and text in a 20-degree field of view.
Source: Mixed News