EVENT
Event News
Talk on " Bridging the Silence: AI-Powered Sign Language Technologies for Inclusive Communication in Southeast Asia" by Dr. Chantri Polprasert (AIT)
We are pleased to inform you about the upcoming seminar by Dr. Chantri Polprasert (AIT) titled:" Bridging the Silence: AI-Powered Sign Language Technologies for Inclusive Communication in Southeast Asia" Everyone interested is cordially invited to attend!
Title:
Bridging the Silence: AI-Powered Sign Language Technologies for Inclusive Communication in Southeast Asia
Abstract:
In Thailand and across much of Southeast Asia, the shortage of qualified sign language interpreters continues to limit access to essential information, education, and public services for the deaf community. This talk presents our ongoing research efforts to address this challenge through the development of AI-driven sign language technologies.
One of our key initiatives is a proposed system for the automatic production of Thai Sign Language (TSL) translations, particularly for use in broadcast media. This project--currently under revision by the National Broadcasting and Telecommunications Commission (NBTC) for funding--aims to enable real-time sign language generation using an animated virtual signer. In collaboration with Thai PBS, we are exploring the development of a sign language avatar application to automatically translate news content into TSL, enhancing media accessibility for deaf viewers. As part of this initiative, we are also developing a Thai Sign Language corpus that captures vocabulary, grammar, and non-manual features necessary for accurate recognition and synthesis.
Expanding beyond Thailand, we are collaborating with Multimedia University (MMU), Malaysia, and Universitas Sumatera Utara (USU), Indonesia, to develop an ASEAN Sign Language Translation System. This system is designed to support multilingual sign-to-text and text-to-sign translation across Thai (TSL), Malaysian (BIM), and Indonesian (Bisindo) sign languages. By leveraging deep learning, gloss alignment, and transfer learning, we aim to address the challenges of linguistic diversity and limited training data in the region.
We also draw on experience in American Sign Language (ASL) recognition, applying both keypoint-based methods with MediaPipe and video-based models such as the Video Swin Transformer. These complementary approaches help us evaluate model robustness and inform the design of sign language recognition systems that are adaptable across languages and contexts.
Although these projects are still in development, their goal is to support real-world applications that help bridge the communication gap for deaf communities--particularly in education, digital accessibility, and public services. This talk will highlight our technical approach, collaborative efforts, and vision for inclusive, AI-powered communication across Southeast Asia.
Speaker Bio:
Dr. Chantri Polprasert is an Assistant Professor in the Department of Information and Communication Technologies at the Asian Institute of Technology (AIT), Thailand. His research focuses on artificial intelligence, machine learning, and data science, with a particular emphasis on sign language intelligence and inclusive communication technologies. He holds a Ph.D. in Electrical Engineering from the University of Washington, Seattle and has received multiple national awards for innovation in AI and socially impactful technologies.
Dr. Polprasert has authored over 30 peer-reviewed publications, including articles in high-impact journals such as IEEE Transactions on Wireless Communications, Signal Processing (Elsevier), Physical Communication, and Journal of Neural Engineering. His interdisciplinary research spans deep learning for sign language recognition, P300-based brain-computer interfaces, anomaly detection in network systems, and underwater acoustic communications. His recent work explores keypoint-based sign language recognition and multilingual translation systems for ASEAN sign languages.
In addition to his academic contributions, Dr. Polprasert actively collaborates with partners across Southeast Asia, including Multimedia University (Malaysia) and Universitas Sumatera Utara (Indonesia), on AI-powered accessibility projects. He also serves as a reviewer for top-tier IEEE journals and conferences and is certified in AWS Cloud Architecture, with hands-on expertise in cloud-based system design and data engineering.
Time/Date:
13:30 - 14:30 May 8 (Thursday), 2025
Place:
Room 1903 , NII and online
Contact:
If you would like to join, please contact by email.
Email :bono[at]nii.ac.jp