Fish Audio raises $50M seed to build AI voice models for creators and enterprises

Fish Audio, founded by former NVIDIA researcher Shijia Liao, has raised $50 million in a seed round led by Coreline Ventures and Capital Today, with participation from other investors. The startup provides a library of over 15,000 natural language controls for AI voice models, catering to both creators and enterprises. Since its launch last year, it has attracted over 8 million users and achieved $21 million in annual recurring revenue. Fish Audio has released five models, including four speech generation and one speech-to-text model, with three open-sourced and the latest S2.1 Pro available via paid API. The company offers paid plans for creators and teams, as well as enterprise APIs used by organizations like HeyGen, Sanas, and Plaud. It faced controversy when creators alleged unauthorized voice uploads, but has since automated a takedown process that removes voices in under three minutes. CEO Rissa Cao stated that the funding will help develop more advanced models, including an audio understanding model and a speech-to-speech model. The voice generation market is competitive, with rivals like ElevenLabs and WellSaid, but investors believe Fish Audio's fine-grained controls and cost-efficient training give it an edge.