Design of digital cooking assistant system with modern voice generative AI model

This paper presents the design and implementation of digital cooking assistant that leverages modern generative AI models to enhance accessibility and inclusiveness in everyday cooking, named as "Cooking Master". The proposed system integrates a voice-based assistant capable of real-time dialogue, recipe personalization, and hands-free control of the application interface. By combining LiveKit for low-latency communication, Groq hardware acceleration, and the LLaMA language model, the assistant enables natural interaction through speech recognition and synthesis. Particular emphasis was placed on creating solutions that support disabled users, especially the visually impaired, for whom traditional recipe formats often pose significant barriers. The assistant provides step-by-step instructions, intuitive voice commands, and accessible navigation, allowing independent cooking without the need for touch-based interaction. The multilingual architecture further extends inclusivity, enabling users from diverse linguistic backgrounds to benefit from AI-driven assistance. Experimental results confirmed that the system responds with minimal latency and natural-sounding feedback, making it suitable for real cooking environments. The developed solution demonstrates the potential of generative AI in supporting not only convenience and personalization but also in promoting digital accessibility for users with special needs.