On January 5th, Zhiyuan Robotics, a leading company in humanoid robots in China, announced a deep collaboration with the general artificial intelligence company MiniMax (Shanghai Xiyu Technology). The two companies will accelerate the practical application of multimodal AI robots through technology integration.

In this collaboration, MiniMax will provide full-process AI technological support for Zhiyuan Robotics, ranging from text to speech (TTS). Unlike traditional mechanical voice, MiniMax has developed a customized "character system" tailored for Zhiyuan Robotics' product positioning. This means that future Zhiyuan robots will not only speak but also have more identity-appropriate intonation and interaction logic, significantly enhancing the immersive experience when users interact with the robot.

To achieve more personalized interactions, the two parties have also built a customized prompt strategy based on this character system. This technology supports generating unique voices for different users, achieving a "different face for each person" speech synthesis effect, thus meeting differentiated interaction needs in various complex scenarios such as home and office.

In addition, Zhiyuan Robotics will also introduce MiniMax's self-developed music generation model, trying to explore more possibilities in entertainment scenarios. As a global leader in general artificial intelligence, MiniMax has already released several multimodal large models, including MiniMax M2.1, Speech2.6, and Music2.0.