IT Home reported on August 16 that Alibaba Cloud's "Magic Community" has now launched Live Portait, which uploads a photo and a text or voice to generate a digital human video of speaking.
IT Home measured that users upload image photos after entering the application interface, and can choose text drivers (within 100 words) or audio drivers, providing 28 sound choices, and if the lips and teeth are not clear, they can choose to turn on the "lip and teeth repair" function. In addition, the user can set the blink frequency, choosing between 0-4, 0 means no blinking.
After the settings are complete, the uploaded image can be broadcast like a video, which can be applied to scenarios such as live video, chatbots, and enterprise marketing. According to reports, Live Portait related technologies have been included in international AI summits such as CVPR and ICCV.
At the beginning of this month, the Modai community launched two open source models, Qwen-7B and Qwen-7B-Chat, which Alibaba Cloud confirmed as a general model and a dialogue model of 7 billion parameters, both of which are open source, free and commercially available. Among them, Qwen-7B is a dock model that supports multiple languages such as Chinese and English, while Qwen-7B-Chat is a Chinese-English dialogue model based on the base model.
According to public information, Modai is the first AI model open source community in China jointly launched by Ali Damo Academy and the Open Source Development Committee of the China Computer Society (CCF) in 2022, opening more than 300 models to AI researchers and teams in China, covering natural language processing, vision, speech, multimodal and other models.