Recently, the 2024 International Information and Communication Exhibition (referred to as the "PT Exhibition"), hosted by the Ministry of Industry and Information Technology, was held at the National Convention Center in Beijing. During the event, the China Academy of Information and Communications Technology (CAICT) officially released the "2024 Outstanding Cases of Edge-side General AI Innovation Applications." Huilan Technology Co., Ltd. was successfully selected for its technological innovation and foresight with the "AI-powered Multi-modal Virtual Digital Human Generation System based on the AiGCP Smart Large Model Platform." Also included in the list were China Mobile, China Telecom, the Civil Aviation Administration of China, the National New Display Technology Innovation Center, and other central state-owned enterprises and national-level technology innovation centers.

Edge-side general artificial intelligence refers to AI systems deployed on terminal devices. Compared to traditional cloud-based AI, edge-side AI features real-time processing, privacy and security protection, low energy consumption, high efficiency, and offline capability. With the improvement of hardware performance and the optimization of AI algorithms, edge-side general AI is becoming an emerging trend.

The virtual digital human generated by Huilan Technology Co., Ltd.'s "AI-powered Multi-modal Virtual Digital Human Generation System based on the AiGCP Smart Large Model Platform" is built upon key technologies and applications of next-generation cognitive intelligence-driven multi-modal natural human-machine interaction. This system combines human-like features to create multi-modal virtual digital humans. These virtual beings exhibit realistic human facial expressions and movements, and can interact through speech recognition, semantic understanding, speech synthesis, and synchronized facial expressions and actions. Additionally, they leverage general or vertical large models to enable real-time intelligent voice interaction.

Virtual digital humans generated using large models can be widely applied in industries such as government services, finance and insurance, medical consultations, and education. For example, in government services, after fine-tuning based on a large amount of textual content, such as service matters, service guides, and policy interpretations, they can respond in real-time through voice Q&A to users' inquiries about government-related matters in areas like education, employment, entrepreneurship, daily life, and public services. They can also provide detailed voice broadcasts, introducing various service requirements, application processes, contact information, and other service-related details in real-time.