Rissa Cao is co-founder and CEO of Fish Audio. Her public writing and interviews focus on expressive voice generation, the product needs behind it and the company’s work with creators and enterprise customers.[1][2]
Start with 5 Models, 22 People, 1 Year →
X · rissa_cao Career
Contributions
The Drama 3 release presents natural-language direction of emotion and delivery, with local changes to a passage instead of regenerating an entire chapter. Cao describes the intended workflow for audiobooks and multilingual dubbing. The model and release belong to the Fish Audio team, with Shijia Liao and Jiahua Liu also named on the publication.[4]
Ideas
In her company essay, Cao distinguishes getting the words right from performing them. Her earlier voice-AI experience at Amazon Alexa and Meta led her to focus on delivery across a longer script, rather than the quality of one isolated sentence. She describes creator usage and preferences as inputs to Fish Audio’s post-training work.[3]
Tools, writing & talks
Writing, research & talks
Start here
5 Models, 22 People, 1 Yearfish.audioA first-person company account of expressive voice generation, user feedback and product development.
Begin with the distinction between reading words and performing them, then the account of user feedback. It needs no model-training background or paid account; its business metrics are the company’s own report.
Author and CEO[3]- Fish Audio Drama 3fish.audio
A team release explaining natural-language control of voice delivery and editing.
Co-founder and CEO; contributor to team publication[4]