TMTPOST— Chinese AI unicorn MiniMax has launched its next-generation voice model, Speech-2, claiming top performance in two of the world’s most authoritative voice AI benchmarks: Artificial Analysis and Hugging Face TTS Arena. According to the company, Speech-2 surpasses models from leading global players including OpenAI and ElevenLabs, securing the No. 1 position on both benchmark leaderboards. The announcement marks a significant step forward for China’s voice AI capabilities, positioning MiniMax as a serious contender in the highly competitive speech technology space. Founded in 2021, MiniMax has quickly emerged as one of China’s most prominent generative AI startups, backed by strong investor interest and a growing suite of multimodal products. The company has not yet disclosed when Speech-2 will be made widely available, but industry watchers see the achievement as a major milestone in China’s race to develop globally competitive foundational models.