Sign Language Translator
Year: 2024
Type: article
Abstract: Sign language translation is necessary to increase inclusivity and communication within the Deaf and hard-of-hearing communities [1]. Communication difficulties with non-sign language speakers affect social, professional, and educational relationships for deaf people [1]. These issues can be solved by utilizing technology, such as sign language recognition, which offers substitute communication channels [2]. This work presents a novel method for recognizing and translating sign language motions using MediaPipe Holistic and Long Short-Term Memory (LSTM) neural networks [3]. After extensive testing, our LSTM-based model shows impressive accuracy, which makes it a useful tool for real-time translation [4]. We discuss problems encountered and their solutions, providing insights into the model architecture, training protocols, and data preprocessing [5]. Our results suggest that systems for translating and recognizing sign language have a lot of room for improvement. This method encourages a thorough assessment of its effectiveness and has potential uses for the Deaf community, sign language interpreters and technology developers. This research lays the foundation for future sign language translation technology developments, closing the communication gap between the Deaf and hearing communities.
Source: Nanotechnology Perceptions
Language: English
Subfield: Human-Computer Interaction
Field: Computer Science
Domain: Physical Sciences
SDG: Reduced inequalities
Open Access status: diamond
OpenAlex ID: W2561568886