Archive — US — techcrunch
Fish Audio raises $50M seed to build AI voice models for creators andenterprises
Fish Audio, a Palo Alto-based startup, raised $50M in seed funding to develop AI voice models for creators and enterprises, catering to diverse needs from expressive content in gaming to realistic customer support. Founded by former NVIDIA researcher Shijia Liao, the company has over 8M users, generating $21M annually, and seeks to enhance voice generation technology while focusing on creator consent and transparency.
Key points
- Fish Audio raised $50M in seed funding led by Coreline Ventures.
- The startup has developed AI voice models aimed at both creators and enterprises.
- It offers over 15,000 language controls and has more than 8M users.
- The company generates $21M in annual recurring revenue.
- Fish Audio started as an open-source project by ex-NVIDIA researcher Shijia Liao.
- The company has experienced issues with voice consent from creators but has since automated its takedown process.
- Future plans include releasing an audio understanding model and a speech-to-speech model.
Why it matters
- The AI voice model market is growing as more companies seek automation solutions.
- Concerns about voice consent and copyright highlight ongoing issues in the AI sector.
- By 2025, trends in IT suggest a continued push for transparency and user consent in AI applications.
Fish Audio's significant funding and user base highlight the growing demand for sophisticated AI voice solutions, while its approach to creator consent reflects a needed shift in the tech industry toward ethical practices.