Fish Audio raises $52 million seed round for AI voice models

AI voice model startup Fish Audio has raised a $52 million seed round led by Coreline Ventures and Capital Today. The company, which generates $21 million in annual recurring revenue from over 8 million users, will use the funds to expand its enterprise offerings. Fish Audio has faced issues with creator voice consent but has since automated its takedown process to address these concerns.

Fish Audio seed funding

• ▪The funding round for Fish Audio was led by Coreline Ventures and Capital Today.
• ▪AI voice model startup Fish Audio raised $52 million in a seed funding round.
• ▪Participants in Fish Audio's seed round included 359 Capital, Parable, Play Time, Alphalist Partners, Bayhouse Ventures, Carya Venture Partners, and HF0.
• ▪Fish Audio sought capital to develop more advanced models and serve enterprise clients as investor interest grew.

Natural language voice controls

• ▪Fish Audio was founded by former Nvidia researcher Shijia Liao, who initially trained a voice-generation model on a single GPU.
• ▪In the last year, Fish Audio has launched five models: four for speech-generation and one for speech-to-text.
• ▪Fish Audio's latest model, S2.1 Pro, is exclusive to its paid API, while three of its speech-generation models are open-sourced.
• ▪Fish Audio's platform provides a library of more than 15,000 natural language controls for its AI voice models.
• ▪Fish Audio plans to release an audio understanding model and a speech-to-speech model in 2026.

Creator voice consent issues

• ▪The company builds its voice library by compensating users who submit their voices for model training.
• ▪Fish Audio automated its DMCA takedown process, enabling voice removal in under three minutes with proof of ownership.
• ▪Osuke Honda of lead investor Coreline Ventures stated that creator trust requires building consent, transparency, and attribution into the product.
• ▪Fish Audio faced allegations that creators' voices were uploaded to its platform without their consent.

Enterprise customer adoption

• ▪Fish Audio offers an enterprise version of its platform, used by companies such as HeyGen, Sanas, and LiveKit.
• ▪Enterprise clients use Fish Audio for varied needs, including realistic AI avatars, expressive game characters, and low-latency voice agents.

Revenue growth metrics

• ▪Fish Audio generates $21 million in annual recurring revenue.
• ▪More than 8 million people use the open source or hosted versions of Fish Audio's models.
• ▪The company's Fish Speech open-source repository on GitHub has over 31,000 stars.

Speech generation model competition

• ▪An investor from 359 Capital, Rico Mallozzi, noted that Fish Audio's competitive advantages include fine-grained developer controls and cost-efficient model training.
• ▪Competitors in the speech-generation market include ElevenLabs, WellSaid, Cartesia, Speechify, Async, and Krisp.

1 source