Fish Audio is an AI voice cloning and text-to-speech platform that lets users create highly realistic voice clones from as little as 10 seconds of audio. It offers a browser-based editor, a public model marketplace where creators share voice models, and a developer API. Fish Audio's open-source Fish Speech model is one of the most capable multilingual TTS models available, supporting English, Chinese, Japanese, Korean, and more. The platform is widely used for dubbing, audiobook narration, content creation, and building voice AI applications.