Google Glass could be making a comeback with the potential help of Project Astra, Google’s new supercharged computer vision AI. While Google has not committed to integrating its prototype AI assistant into any hardware just yet, a designer has taken matters into their own hands in the meantime.
Pietro Schirano, a designer with experience at Uber and Facebook, has created DIY-Astra, a chatbot with vision capabilities built using Google’s Gemini generative AI. Although not as advanced as Google’s Project Astra or OpenAI’s GPT-4o, DIY-Astra offers a glimpse into the future of improved AI chatbots.
Schirano utilized Gemini 1.5 Flash, Google’s lightweight AI model designed for speed and efficiency, to develop DIY-Astra. By tapping into a live webcam feed, the chatbot can analyze and generate text responses based on what it sees, even reading them aloud. Impressively, DIY-Astra can identify objects in the camera feed and solve handwritten math problems quickly and accurately.
While DIY-Astra is available now through Schirano’s GitHub repository, users may need coding skills to build it. Future plans for the chatbot include adding voice command capabilities to make it even more similar to Project Astra.
As for Google’s Project Astra, the tech giant has showcased its capabilities through smartphone camera demos. While there is no official release date for the prototype assistant, Google has hinted at integrating Project Astra into its products later this year. Until then, DIY-Astra provides a taste of what’s to come in the world of AI chatbots.
