American sign language is the third most common language in the United States, but AI tools developed with ASL data are far fewer than data representing the country’s most common languages, English and Spanish.
Nvidia, American Association of the Deaf and Creative Agency Hello Monday, will help bridge this gap with Signs, an interactive web platform built to support ASL learning and the development of accessible AI applications. It’s there.
Sign language learners access the platform’s library of verified ASL signs, expand their vocabulary with the help of a 3D avatar showing the sign, and analyze webcam footage to receive real-time feedback on signatures. You can use the tool. Signers of any skill level can contribute by signing specific words to help build open source video datasets for ASL.
The dataset NVIDIA aims to grow into 400,000 video clips representing 1,000 signed words, is verified by fluent ASL users and interpreters to ensure the accuracy of each signature and provide high quality visualization It brings you a dictionary and educational tools.
“Most deaf people are born to hear their parents. By providing accessible tools such as signs to begin early learning in ASL, it is effective with children between 6 and 8 months of age. It allows us to open communication channels,” said Cheri Dowling, executive director of the American Association of Hearing Impaired Children. “And knowing that professional ASL teachers have validated all the vocabulary on the platform will give users confidence in what they are learning.”
The NVIDIA team will use this dataset to develop further AI applications that will break down barriers to communication between the hearing impaired and the hearing community. This data will be made public as a resource for building accessible technologies such as AI agents, digital human applications, and video conferencing tools. It can also be used to enhance signage and enable ASL platforms across the ecosystem with real-time AI-driven support and feedback.
Supporting ASL education and exploring language nuances
During the data collection phase, the signs already provide a powerful platform for ASL language acquisition, providing individuals with opportunities to learn and practice an initial set of 100 signs, and with friends and family who use ASL. You can communicate more effectively.
“The sign learning platform helps families with hearing impaired children quickly search for specific words and create corresponding signs. This is the daily life of ASL outside of more formal classes. “It’s a useful tool to support the use of the software,” Dowling said. “We see both the kids and parents exploring it, and I think they can play together.”
The signs currently focus on the hand movements and finger positions of each sign, but the ASL also incorporates facial expressions and head movements to convey meaning. The team behind the sign is to investigate how these non-manual signals can be tracked and integrated into future versions of the platform.
It also explores how other nuances, such as regional variations and slang terms, can be expressed in signs to enrich the ASL database. Signage platform experience for hearing impaired users.
“Improved accessibility at ASL is a continuous effort,” said Anders Jessen, founding partner at Hello Monday/Dept. alphabet. “The signs help to the need for advanced AI tools that help transcend barriers to communication between the hearing impaired and the hearing community.”
The dataset behind the sign is expected to be released later this year.
Start learning or contributing Signs at Signs-Ai.com and learn more about Nvidia’s trusted AI initiatives. Participants at Nvidia GTC, a global AI conference held in San Jose from March 17th to 21st, will be able to participate in autographs live at the event.
