Fish Audio Launches Drama 3: Direct AI Voices in Plain Language
Fish Audio has released Drama 3, an AI model enabling control over synthesized voices using plain language. The new model offers creators more precise command over voice performance editing.
Palo Alto, Calif. – Fish Audio has launched Drama 3, a new text-to-speech model designed to give developers, studios, and creators finer control over AI-generated voice performances. The model allows users to direct voice acting using natural language prompts, adjusting tone, pacing, and character emotion.
Drama 3 enables users to describe desired vocal characteristics and emotions in simple terms, with the ability to modify performances line by line. This feature is particularly beneficial for applications such as audio dramas, audiobooks, game development, and animation, where nuanced vocal direction is crucial for impactful storytelling.
"Voice AI has become very adept at generating speech. The next step is directing a performance," said Rissa Cao, co-founder and CEO of Fish Audio. She highlighted that Drama 3 places creators in the role of director, allowing them to refine vocal output without needing to re-render entire sections.
The model introduces four key capabilities: natural language direction, multi-speaker scene generation, audio inpainting for segment editing, and consistency for long-form and multilingual content. Fish Audio has recently experienced significant growth in user base and revenue, with its technology being adopted across various content creation sectors.