Fish Audio seed funding reaches $52 million for AI voice platform
Fish Audio raised a $52 million seed round as the voice AI startup pushes beyond text-to-speech into enterprise audio tools.
By Dominic Okoye · Staff Writer
· 3 min read
Fish Audio seed funding has reached $52 million, with Coreline Ventures and Capital Today leading the round for the AI voice startup legally known as Hanabi AI Inc. The company said it will use the capital to expand from text-to-speech and voice cloning into a broader audio-native stack for developers and enterprises.
The round also included 359 Capital, Play Time, HF0, 645 Ventures, Parable, Carya Venture Partners, Alphalist Partners and unnamed angel investors. Fish Audio did not disclose a valuation. For a seed-stage company, the round is large, and the disclosed operating metrics explain part of the investor appetite: Fish Audio says it has more than 8 million users and annual recurring revenue above $21 million.
What does Fish Audio do?
Fish Audio sells voice AI infrastructure for developers building text-to-speech, voice cloning and voice agent products. Its pitch is that generated voices can be controlled at the word level for emotion, tone, inflection and pacing, using more than 15,000 natural-language prompts.
The company traces its product to co-founder and Chief Scientist Shijia Liao, a former Nvidia video researcher who began training voice models on a single laptop GPU. That work became Fish Speech, an open-source project that Fish Audio says has drawn more than 31,000 GitHub stars since the company’s 2023 launch.
Fish Audio says its platform can clone a voice from a five-second recording in under 15 seconds and supports 83 languages. It also says its S2.1 Pro model beat rival systems in blind listening tests, with 67% of listeners choosing Fish Audio outputs over alternatives. Those are company-reported benchmarks, and the usual caveat applies: blind tests vary by prompt set, language, listener pool and production constraints.
The early customer base centered on creators, game developers and indie builders, according to the company. Fish Audio now says it is selling into regulated sectors including healthcare and financial services, where it offers on-premises deployment, zero-data retention and HIPAA compliance for customers that cannot send sensitive audio to a vendor-hosted service.
How Fish Audio plans to use the money
Fish Audio said the new funding will support development of voice-native large language models and real-time speech-to-speech translation tools. The company also plans to build out enterprise sales and add more API integrations, including through partners Retell AI and LiveKit.
One near-term distribution move is a free developer push. Fish Audio said it will make its S2.1 Pro model available at no charge through its official API starting at the end of August. The company did not specify usage limits or how long that free access will last.
The funding puts Fish Audio in the same conversation as better-capitalized voice AI companies, especially ElevenLabs, which recently raised $500 million at an $11 billion valuation. Fish Audio is much earlier by financing stage, but its reported ARR gives investors a clearer baseline than many AI infrastructure startups offer at seed.
Chief Executive Rissa Cao said the company’s goal is to make human-sounding AI voices available across users ranging from beginner creators to large enterprises. Coreline Ventures managing partner Osuke Honda said the investment reflects a view that voice is becoming a default interface for AI systems, citing Fish Audio’s performance, multilingual support, emotional controls and cost as reasons for backing the company.
This story draws on original reporting from SiliconANGLE.
