Croco is a real-time conversational artificial intelligence (AI) avatar system that integrates Automatic Speech Recognition (ASR), Large Language Models (LLMs), and Text-to-Speech (TTS) technologies into a unified framework. By incorporating Semantic End-of-Turn Detection alongside Voice Activity Detection (VAD), the system facilitates natural, low-latency, and autonomous dialogue without the need for manual intervention (i.e., hands-free operation). Furthermore, the architecture utilizes a 3D avatar featuring real-time, synchronized lip-syncing and contextual gestures, and is specifically optimized to deliver accurate and instantaneous responses to inquiries concerning King Mongkut's Institute of Technology Ladkrabang (KMITL).