Fish Audio (Hanabi AI Inc.) raises $52M in Seed

Fish Audio, officially known as Hanabi AI Inc., has secured $52 million in seed funding to advance its voice-based artificial intelligence services. The fundraising effort was led by Coreline Ventures and Capital Today, with additional backing from investors including 359 Capital, Play Time, and several others. This significant influx of capital underscores the startup's ambition to position voice as the default interface for AI models, a vision spearheaded by its co-founder and Chief Scientist, Shijia Liao.

The origins of Fish Audio trace back to Liao's frustration with the monotonous synthetic voices of early AI technologies. This led him to develop more dynamic and expressive voice AI models from his laptop. The result was an open-source project that rapidly gained traction within the developer community, attracting attention from indie creators and gaming designers. Fish Audio’s platform has since evolved, featuring capabilities such as fine-grained emotion control and the ability to clone voices from brief audio clips. The platform supports 83 languages, and in blind listening tests, 67% of participants favored its outputs over competitors.

Strategically, Fish Audio aims to extend its reach beyond its initial focus on video game developers and content creators, targeting regulated sectors including healthcare and financial services with secure, compliant deployments. The recent funding round enhances Fish Audio’s competitive positioning against larger entities like ElevenLabs Inc., which recently raised $500 million at a substantial valuation. CEO Rissa Cao emphasizes the democratization of high-quality, human-sounding AI voices, serving a diverse user base ranging from amateurs to large enterprises, while avoiding customer privacy issues through zero-data retention solutions.

In a market where voice AI capabilities are becoming increasingly pivotal, Fish Audio's focus on customizable, human-like interaction sets it apart from incumbent players. The firm's rapid adoption—boasting over eight million users and generating an annual recurring revenue of over $21 million—is indicative of strong demand for enriched voice interfaces. Fish Audio's commitment to making conversation with AI models more natural and reliable could lead to broader adoption across varied industries.

Looking ahead, Fish Audio plans to expand its offerings beyond text-to-speech technologies, by building a comprehensive audio-native stack incorporating large language models optimized for voice interaction. This advancement will likely pave the way for new applications and opportunities within the voice AI space. The success of these expansions will depend on continuous innovation and the ability to maintain its user base while navigating potential regulatory landscapes.