AI voice startup Fish Audio has secured $50 million in seed funding led by Coreline Ventures and Capital Today. The company, which provides expressive voice models for creators and enterprises, has reached $21 million in annual recurring revenue in its first year. Investors are now watching how the firm scales its API business while managing ongoing concerns regarding voice ownership and data misuse.
Detailed Coverage
Fish Audio, a startup focused on artificial intelligence voice models, has raised $50 million in a seed funding round led by Coreline Ventures and Capital Today. The company intends to use this capital to further develop its technology, which offers expressive and steerable AI voices designed for both creative professionals and enterprise applications.
Rapid Financial and User Growth
Launched just last year, Fish Audio has reported reaching 8 million users. The company has also achieved $21 million in annual recurring revenue, a common industry metric used to measure the predictable portion of a subscription-based business’s income. The platform provides a mix of open-source models and a paid, advanced version known as S2.1 Pro, which is accessible through an application programming interface (API). This tiered structure allows the company to serve independent developers and larger organizations alike, with current clients including HeyGen, Sanas, and Plaud.
Origins and Competitive Landscape
The startup was founded by Shijia Liao, a former researcher at NVIDIA. The company initially gained traction within the developer community by open-sourcing its speech generation models on GitHub, where its primary repository has secured over 31,000 stars. Fish Audio is entering a competitive sector that includes established players like ElevenLabs and WellSaid. To differentiate itself, the company is prioritizing fine-grained controls for developers and cost-efficient model training, which it hopes will help it maintain an edge against larger, well-funded incumbents.
Managing Voice Ownership Risks
The growth of AI voice technology has brought significant attention to the risks of unauthorized voice cloning and misuse. Fish Audio has implemented an automated process to handle Digital Millennium Copyright Act (DMCA) takedown requests, allowing content creators to prove ownership and request the removal of their voice data from the platform in under three minutes. This move is part of a broader industry push for transparency, with investors like Oskue Honda of Coreline Ventures highlighting that future success in this space will depend on establishing trust and potentially implementing revenue-sharing models for voice contributors.
Looking ahead, Fish Audio plans to launch an audio understanding model and a speech-to-speech model later this year. Investors and industry observers will be watching the company's ability to maintain its revenue growth while navigating the evolving regulatory environment surrounding synthetic media and intellectual property rights.