Fish Audio, an artificial intelligence startup officially known as Hanabi AI Inc., has raised $52 million in seed funding led by Coreline Ventures and Capital Today, with participation from 359 Capital, Play Time, HF0, 645 Ventures, Parable, Carya Venture Partners, Alphalist Partners, and unnamed angels. Co-founder and Chief Scientist Shijia Liao, a former video researcher at Nvidia Corp., began training voice AI models with a single graphics processing unit in his bedroom, developing the open-source Fish Speech project that amassed more than 31,000 stars on GitHub. Since its launch in 2023, Fish Audio has evolved into a voice AI platform providing word-level emotion controls driven by over 15,000 natural language prompts. It claims to clone a voice from a five-second audio clip in less than 15 seconds, supports 83 languages, and its flagship model S2.1 Pro outperformed competitors in blind listening tests, with 67% of listeners preferring its outputs. The platform’s user base has grown to over eight million, and its annual recurring revenue exceeds $21 million.