Fish Audio Raises $52M Seed for AI Voice Model Development

Funding Details

Fish Audio announced it has raised $52 million in a seed funding round. The round was led by Coreline Ventures and Capital Today, with participation from 359 Capital, Parable, Play Time, Alphalist Partners, Bayhouse Ventures, Carya Venture Partners, and HF0.

AI Voice Model Offerings

Fish Audio develops AI voice models designed for both creative applications requiring expressiveness and enterprise uses needing steerability for customer support and sales operations. The company's library includes over 15,000 natural language controls for its voice models.

Growth and Adoption

Since its launch last year, Fish Audio has accumulated over 8 million users across its open-source and hosted model versions. The company reports an annual recurring revenue of $21 million. It has released five models in the past year, including four speech generation models and one speech-to-text model. Three of its speech generation models are open-source, while the S2.1 Pro model is available via a paid API.

Enterprise and Creator Solutions

Fish Audio offers paid monthly plans for creators and teams, which include generation minutes and voice cloning features. An enterprise version of its APIs and platform is also available, with companies like HeyGen and Sanas utilizing its services. The company notes that different enterprises have varying needs, such as realism for AI avatars, expressiveness for gaming characters, or natural-sounding, low-latency voices for calls.

Origin and Challenges

The company originated from a project by former NVIDIA researcher Shijia Liao, who developed a voice generation model due to dissatisfaction with existing synthetic voices. This model was open-sourced, leading to the Fish Speech repository on GitHub, which has over 31,000 stars. Fish Audio previously faced issues regarding user-submitted voices, with some creators alleging their voices were used without consent, which the company addressed through a DMCA takedown process.