NVIDIA’s AI-Powered Signs Platform Expands ASL Learning & Accessibility

American Sign Language (ASL) is the third most widely used language in the United States, yet AI tools built with ASL data remain far less common than those for English or Spanish. To help close this gap, NVIDIA, the American Society for Deaf Children, and creative agency Hello Monday have launched Signs, an interactive web platform designed to:

• Support ASL learners with a validated sign language dictionary and real-time feedback.

• Build a high-quality ASL dataset for developing accessible AI applications.

• Advance AI-powered communication tools for the deaf and hearing communities.

An AI-Powered Learning Platform for ASL

The Signs platform offers an engaging way to learn ASL, featuring:

• A validated ASL library – Users can explore a growing dictionary of signs, demonstrated by a 3D avatar to ensure accuracy.

• AI-assisted feedback – A built-in AI tool analyzes webcam footage to provide real-time guidance on signing accuracy.

• A crowdsourced ASL dataset – Signers of all skill levels can contribute by recording specific words, helping to build a robust dataset for AI-driven accessibility tools.

To ensure accuracy and inclusivity, all signs in the dataset are validated by fluent ASL users and interpreters. NVIDIA’s goal is to grow the dataset to 400,000 video clips, covering 1,000 signed words—creating a high-quality visual dictionary and teaching tool.

Enhancing ASL Accessibility With AI

Beyond ASL education, NVIDIA plans to use this dataset to develop AI applications that bridge communication gaps between deaf and hearing communities. The dataset will be publicly available to support accessible technologies, such as:

• AI-powered sign language agents

• Digital human applications

• Video conferencing tools with ASL recognition

Additionally, Signs could be further enhanced with AI, allowing real-time feedback to evolve into a broader ASL learning ecosystem.

Expanding ASL Learning and Cultural Nuances

Currently, Signs teaches hand movements and finger positions, but ASL also relies on facial expressions and head movements to convey meaning. The team is actively researching ways to track and integrate these non-manual signals in future updates.

Other areas of exploration include:

• Regional variations and slang – Ensuring ASL nuances are represented in the platform.

• User experience research – In collaboration with the Rochester Institute of Technology’s Center for Accessibility and Inclusion Research, Signs is being evaluated for usability and effectiveness among deaf and hard-of-hearing users.

“Improving ASL accessibility is an ongoing effort,” said Anders Jessen, founding partner at Hello Monday/DEPT. “Signs can serve the need for advanced AI tools that help transcend communication barriers between the deaf and hearing communities.”

Get Involved With Signs

The Signs dataset is expected to be released later this year, providing an open resource for researchers and developers working on ASL-based AI applications.

• Start learning or contributing at signs-ai.com.

• Experience Signs live at NVIDIA GTC, a global AI conference from March 17-21 in San Jose.

Looking Ahead

As AI-powered tools like Signs continue to evolve, they have the potential to transform ASL education and accessibility. By integrating facial expressions, regional variations, and advanced AI recognition, future versions of Signs could provide even more accurate and personalized learning experiences.

Beyond education, this technology could play a key role in breaking down communication barriers, enabling real-time ASL translation in video calls, public spaces, and workplaces. With NVIDIA’s commitment to open-access datasets, researchers and developers can leverage this resource to build new AI-driven tools that enhance inclusion for the deaf and hard-of-hearing communities.

As ASL AI continues to advance, Signs represents a major step toward a world where technology enables truly accessible communication for all.

Editor’s Note: This article was created by Alicia Shapiro, CMO of AiNews.com, with writing, image, and idea-generation support from ChatGPT, an AI assistant. However, the final perspective and editorial choices are solely Alicia Shapiro’s. Special thanks to ChatGPT for assistance with research and editorial support in crafting this article.