VideoChat is a real-time voice interaction digital human project that supports end-to-end voice solutions (GLM-4-Voice - THG) and cascading solutions (ASR-LLM-TTS-THG). Users can customize the appearance and voice of the digital human, with voice cloning capabilities that require no training, achieving initial package latency as low as 3 seconds. This project leverages the latest AI technologies, including Automatic Speech Recognition (ASR), Large Language Models (LLM), End-to-End Multimodal Large Language Models (MLLM), Text-to-Speech (TTS), and Talking Head Generation (THG), to provide users with a highly customizable and low-latency interaction experience.