With the continuous development of artificial intelligence technology, virtual image, virtual anchor and other applications have emerged in large numbers, and application scenarios are also very wide, such as: virtual customer service robot, virtual host / anchor, virtual image / virtual post, virtual image VR / AR, digital twin.
Beijing zhongkehuilian Technology Co., Ltd. (hereinafter referred to as "Huilan Technology") and the National Engineering Laboratory of Beijing University of science and technology cooperate in depth to realize the image of virtual intelligent human, plus intelligent multimodal voice interaction, and realize the perfect combination of AI + virtual human.
The solution of Huilan Technology virtual intelligent human platform supports a variety of facial expressions, and has 2D / 3D virtual image, which can be customized with real person image or 3D virtual image.
Based on speech recognition, semantic understanding and other technologies, the solution of Huilan Technology virtual intelligent human platform realizes real-time analysis of voice and audio information.
Based on mature speech synthesis technology, skeleton and action specification, the solution of Huilan Technology virtual intelligent human platform customizes personalized speech database for virtual intelligent human, and realizes the action of the model through voice content.
Based on speech recognition, semantic understanding, speech synthesis and other technologies, the solution of Huilan Technology virtual intelligent human platform realizes intelligent voice interaction and supports multiple rounds of dialogue and interposition.
The solution of CCHP virtual intelligent human platform supports a variety of oral patterns to achieve the fluency and matching of pronunciation between words.
The solution of Huilan Technology virtual intelligent human platform supports personalized customization services, such as virtual image background, image accessories, clothing, etc.