This project presents a prototype web-based interactive platform that integrates textual prompts and AI-generated videos to engage users in speech tasks, with a focus on accessible interaction for children with autism spectrum disorder (ASD).
videos/: AI-generated videos created from prompt texts using a text-to-video model.autism_intervention_prompts.txt: Intervention prompts tailored for speech development in children with autism.prompts_en.txt: General English prompts used in speech tasks.prompts_en_xiaobai.txt: English prompts with name Xiaobai.index.html: Main webpage for launching the interactive task interface.
This GitHub repository powers the website interface available at:
👉 https://wailywang.github.io/autism_speech_task
The videos/ folder contains synthetic videos generated from the text prompts in this repository. These are created using the Wan2.1 T2V-1.3B model—a state-of-the-art text-to-video framework. Each video corresponds to a specific prompt and is designed to provide engaging visual context for children during speech practice.