Creating 3D Objects in VR With Voice Commands

AR/VR Developer at Microsoft Research Jasmine Roberts has recently demonstrated an incredible setup created for the upcoming NeurIPS 2022 conference. Powered by Unity and artificial intelligence, the system allows one to create and place 3D objects in a VR space in real-time with nothing but voice commands, telling the AI what object to place and where. According to Jasmine, the 3D assets seen in the demo are created, loaded, and manipulated via OpenAI's Generative Pre-trained Transformer 3(GPT-3) 

You can check out the developer's original post here. Also, don't forget to join our Reddit page and our Telegram channel, follow us on Instagram and Twitter, where we share breakdowns, the latest news, awesome artworks, and more.