Word to World

Xinyue Yang

Mai / May - 2020

Weißensee Kunsthochschule Berlin

Master of Arts


Draw by speaking? "Word to World" is software that uses speech recognition and natural language processing to transform real-time speech into dynamic visual worlds. It enables the creation of complex imaginary worlds that follow the narrator and the creation of complex animations controlled by linguistic structures.

What is the Topic?

The theme uses machine learning to transform real-time language into animated worlds. Users can create a dynamic visual world just by speaking.

Why does it look like this?

While the user speaks and the computer generates visual elements, the software provides simply designed interfaces that can assist users in speech interactions with the computer.

What is special?

For storytellers who do not have many drawing or animation skills, this software can revolutionise the storytelling and animation industry.

What is new?

In this project, the cutting-edge technology of "natural language processing" has been combined into a useful application. It shows the possibility of a playful collaboration between man and machine. We have mastered intuition, language, imagination and creativity, while the computer has mastered calculation, algorithm, logic and memory. "Word to World" has created a mixture of intuition and logic because we and the computer have a common goal: to tell a story.