Director, Institute for Language and Speech Proceeding (ILSP),
Athena Research Center, Athens, Greece
Title of the talk: Enhancing Accessibility to Tangible and Intangible Cultural Heritage Content
In this presentation, we will discuss two recent research directions aimed at improving accessibility to both tangible and intangible cultural heritage content. The first direction focuses on the development of smart exhibits, specifically designed to cater to visually impaired museum visitors. These smart exhibits are created through a combination of 3D digital representations of museum artifacts or historical monuments and 3D printing. The 3D digital models are meticulously processed to enhance the haptic experience, considering morphological, topological, and dimensional features. The exhibits are equipped with hardware components that allow pairing with smart/wearable devices and communication with a central control point as well as event triggers that activate narration of contextual information. The second research direction concentrates on technologies aimed at improving access to theatrical performances for individuals who are visually impaired, Deaf or Hard of Hearing. The core concept revolves around a real-time "metronome" system, which utilizes Speech-To-Text and NLP technologies to process the actors' speech and align it with the speech acts from the script. This metronome system facilitates the synchronization of multiple information streams, including supertitles, dubbing, audio description, and sign language videos. By combining these elements, the accessibility of the performance is significantly enhanced, allowing individuals with different accessibility needs to fully engage with the theatrical experience.
Institute of German Sign Language and Communication of the Deaf,
Universität Hamburg (UHH), Germany
Title of the talk: Sign Language Technologies and Accessibility
The visual modality, no established writing systems, and data scarcity are the most prominent challenges sign language technologies have to deal with. At the same time, they are key for improving many of the accessibility issues deaf sign language users are confronted with. We review the state of the art in sign language technologies and describe how current progress pushes the limits in sign language accessibility and what is still missing in order to make new language technologies available for the sign language communities around the world.
Department of Electrical Engineering, Computer Engineering, and Informatics,
Cyprus University of Technology (CUT), Cyprus
Title of the talk: Promoting Sustainable Development Goals via Deep Learning for Signal Processing and Understanding
In this talk, we present our latest research in leveraging Deep Learning to promote the achievement of Sustainable Development Goals, and especially Inclusion. In this context, we discuss our work on developing speech-to-text/text-to-speech modules tailored to the requirements of a system that addresses the communication between deaf and hearing people. Additionally, we address the technological challenge of sign language (SL) transcription and generation; the aim is to offer real-time translation of SL video to text or speech and vice-versa, with satisfactory accuracy, on commodity mobile devices such as smartphones and tablets. Finally, we discuss extensions of our work on leveraging Deep Learning to implement an emergency contingency scenario.
Archaeological Museum of Thessaloniki (AMTH), Greece
Title of the talk: Sensitive Museums: Engaging Deaf Audiences with the SignGuide Application
Museum policies strive to enhance accessibility conditions and create inclusive environments for diverse audiences, including those with sensory disabilities. In this regard, the Archaeological Museum of Thessaloniki has undertaken several initiatives, leveraging cutting-edge technologies, to meet the needs of visitors with sensory impairments. This presentation highlights the exemplary practices implemented by the museum, particularly focusing on its involvement in the SignGuide project—a collaborative and multidisciplinary program that resulted in the development of an interactive museum guide in Sign Language.