Fish Audio has announced the release of Drama 3, an innovative text-to-speech model that boasts 35 billion parameters, making it one of the most advanced models in the field. This mixture-of-experts model utilizes approximately 3 billion active parameters to generate high-quality speech from plain language input.
Drama 3 supports a wide array of 83 languages, catering to a global audience and enhancing accessibility in various applications. The model’s open weights are provided under a revenue-capped community license, promoting collaborative development and use within the AI community.
The introduction of Drama 3 underscores the growing demand for sophisticated text-to-speech solutions across industries, from entertainment to education, and highlights Fish Audio’s commitment to pushing the boundaries of AI-driven audio technology.
Compiled automatically by the Tech AI Newsdesk from public AI-news sources and summarised in our own words.
Frequently asked questions
What is Drama 3?
Drama 3 is an innovative text-to-speech model released by Fish Audio, featuring 35 billion parameters.
How many languages does Drama 3 support?
Drama 3 supports a wide array of 83 languages, catering to a global audience.
What type of license does Drama 3 use?
The model's open weights are provided under a revenue-capped community license.
