Fish Audio
fish.audioX / TwitterLinkedIn

AI text-to-speech and voice cloning studio for content creators, developers, and studios

Founders
Rissa CaoShijia Liao
645 Deal Lead
Parul Singh
Industry
AI Infra / DevTools
Entry Stage
Seed
Status
Active
Investment Date
2026
Company Overview

Fish Audio is an AI voice technology platform developed by Hanabi AI Inc. that offers text-to-speech, voice cloning, and speech-to-text services, supporting over 30 languages and a library of more than 2 million voices. The platform provides developer APIs and tools for a range of use cases including audiobook narration, video voiceovers, character voices, and conversational chatbots.

Investment Rationale

Co-founder and Chief Scientist Shijia Liao, a former NVIDIA video researcher, trained his first voice models on a single GPU and turned them into Fish Speech, an open-source project that gained rapid traction on GitHub. That shows rare research talent and capital efficiency. CEO Rissa Cao leads commercialization. The company’s early traction is exceptional: it grew from a side project to $21M ARR and 8 million users in its first year. Voice is becoming the default interface for AI systems, and developers need expressive, multilingual, low-cost voice at scale. Fish Audio’s open-source community gives it a distribution advantage that feeds its paid API, and its models already power AI avatars at HeyGen and organizations like Sanas. Planned expansion beyond text-to-speech into the full audio-native stack, including voice-native LLMs and speech-to-speech, positions it to become core voice infrastructure.