Author丨Huang Yanbing

Leifeng.com·Jingxi exclusively learned that Xiao Yuwei, a former senior algorithm engineer at Honor, has left his job to start a business and founded the AI ​​interactive hardware company KnowVoice in the middle of this year. It is currently being incubated by Shenzhen Science and Technology Innovation Institute under Professor Li Zexiang and has received early investment support.

Founder and CEO Xiao Yuwei was born in 1998. He graduated from Guangdong University of Technology and co-created DynamicX, the Robomaster team of Guangdong University of Technology. After graduation, he worked for Honor, and his career experience spanned low-level technology and strategic business.

Xiao Yuwei worked as a GPU-Turbo-X SE system engineer in his early days, responsible for rendering technology and end-side high-performance optimization. The AI ​​rendering separation project "Honor Phantom Engine" he led mainly solved the problem of occasional frame drops and sudden freezes in the game. In addition, he holds dozens of domestic and foreign patents and represents Honor in the formulation of standards in the graphics industry.

Since then, Xiao Yuwei was transferred to the position of researcher in the President's Office, turning to the direction of games and AI terminal equipment, leading research on game players in the Z era, terminal game experience, and the future development of AI terminals, reporting directly to CEO Li Jian.

It is understood that the core team of the project has been formed, with members coming from Huawei, Tencent, OPPO and other companies and universities.

Li Jiawei, the head of product technology, is a former OPPO senior radio frequency engineer with experience in product interaction design and mass production. He led the smartphone industry's first "buttonless AI intelligent edge interaction"; Lin Shenghui, the head of embedded software, was responsible for embedded project development at Huawei, MEAN WELL and other companies. Dr. Li Hongzhe, the head of the algorithm, graduated from Harbin Institute of Technology with a bachelor's degree and a Ph.D. from Korea University. He studies AI software and system security, context management and memory management, and has served as a senior algorithm engineer at Huawei and Tencent.

The first product launched by the KnowVoice team is a terminal-side input device in the form of a pendant, focusing on understanding personal intentions and assisting in expression.

According to reports, the product does not replace the original input methods of mobile phones, computers and other devices, but is integrated into different applications in a suspended form: whether it is WeChat chat or document writing, users can supplement the content by calling the information previously accumulated by the device.

This is also the different thinking between KnowVoice and current mainstream AI memory hardware.

At present, KnowVoice has a product prototype, and the team uses the Vibe Coding crowd as an early entry point. It is mainly targeted at product developers, hardware engineers, structural engineers, self-media bloggers, independent OPC entrepreneurs and other professional users who need high-frequency expression and information processing.

In terms of data processing, the team chose to keep the core data on the device side, without uploading it to the cloud or passing through the company's servers, hoping to return data control to users.

Leifeng.com (public account: Leifeng.com)·Jingxi will continue to pay attention to the team's subsequent progress. Welcome to add the author WeChat frenchfries0016 to communicate.

Original articles from Leifeng.com, unauthorized reproduction is prohibited. See the reprint instructions for details.

edit

Resend Email Modify Email