Abstract
This technology automates accessibility testing of user interfaces (UIs) by using an artificial intelligence (AI) model to validate screen reader output. The framework uses automation tools to navigate a UI, such as a web page or application, while recording the audio output generated by a screen reader. This initial recording establishes a baseline recording. For subsequent regression testing, the system runs the automation again and compares the new output to the baseline recording to detect any deviations. An AI tool performs the validation by either converting the audio to text for comparison or conducting a direct audio and video comparison. When the system detects a difference between the current output and the baseline, it identifies a potential bug. This automated approach is platform-agnostic and removes the need to create multiple baseline files for different screen reader and platform combinations.
Keywords: automated accessibility testing, screen reader output, speech to text conversion, golden file, artificial intelligence (AI) powered testing, UI automation, video comparison, regression testing
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Jain, Shubham, "AI-Powered Validation of Screen Reader Output for Automated Accessibility Testing", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/10415