Show HN: Parlor Jarvis – Realtime AI (audio+screen in, voice out) & multilingual (github.com)

Parlor Jarvis has been announced as an innovative, on-device multimodal AI that allows users to have natural voice and vision conversations. This enhanced version of the original Parlor system features improved multilingual capabilities and can handle a wider range of inputs, such as camera feeds, screen sharing, PDFs, and videos. With the updated language model Supergemma 4 E4B and the multilingual text-to-speech engine Supertonic, users can interact in five languages including English, Korean, Spanish, Portuguese, and French, all while processing everything locally on their machines. This development is significant for the AI/ML community as it pushes the boundaries of what is achievable with on-device AI, emphasizing real-time processing without reliance on cloud servers. This shift not only reduces latency and cost but also increases accessibility for users looking to learn new languages or seek real-time assistance. Key technical enhancements incorporate voice activity detection, responsive dialogue management, and integration of advanced image understanding capabilities, showcasing the rapid advancement in on-device machine learning technologies and paving the way for more robust, interactive AI applications in everyday scenarios.