‍

We are excited to share that Eminence Ventures has led the inaugural funding round for Yusheng Yueban, a startup pioneering the next generation of emotional voice interaction. Their mission is simple but ambitious: to make conversations with AI sound natural, expressive, and truly human—far from the robotic, delayed voices most of us know today.

Yusheng Yueban was founded by Professor Yanmin Qian, a leading scholar in auditory AI at Shanghai Jiao Tong University, whose research in speech recognition, audio generation, and emotional interaction has already been widely applied across industries. He is joined by Mei Jie, a serial entrepreneur with deep commercialization experience, who ensures that cutting-edge research can quickly translate into market-ready products.

Unlike most voice AI systems that stitch together multiple modules—causing delays and flat, mechanical output—Yusheng Yueban’s Luna-1 model is fully end-to-end. It processes speech from input to response in a seamless loop, enabling conversations that are faster, more fluid, and emotionally engaging. On industry benchmarks, Luna-1 has already demonstrated performance on par with the world’s best, ranking just behind GPT-4o-Audio.

The potential applications are broad, from AI voice assistants and real-time translation to education content, podcasts, virtual avatars, and immersive audio experiences. The team is already working with global leaders like Samsung, Ant Group, Meituan, and Tencent—laying the groundwork for global adoption.

As Eminence Ventures Partner Laven Luo noted: “The Yusheng Yueban team combines world-class technical depth with sharp commercial instincts. Their model delivers lower latency, higher accuracy, and more expressive voices than traditional approaches, while also being deployable on edge devices. This creates a clear path to global commercialization.”

We believe Yusheng Yueban represents an important step forward in making AI interactions more human, and we are proud to support their journey toward global impact.

‍