Understanding hand sign language in human-computer interaction

• Main Author: Er, Alvin Beng Kiong
• Other Authors: Yap Kim Hui
• Format: Final Year Project
• Language: English
• Published: 2009
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/15764
• Institution: Nanyang Technological University

In this project, an English to American Sign Language translator cum American Sign Language learning application, Let’s Talk, had been successfully developed using the main character as Sydney, the 3-D human figure. The objective of this application is to use as a communication tool for the user to communicate with a hearing impaired person. The word or sentence for translation can be typed on the GUI using the keyboard or a speech recognition technology. Quiz is also implemented as a learning tool for the user. Before the design and construction of the software, the software process model was derived and the Incremental Model was chosen. Following the flow of the Incremental Model, the project was divided into two phase. The first is the prototype phase and the second is the enhancement phase. During the software development, Visual Basic 6.0 had been used for programming and GUI creation. The GUI is designed to be user-friendly and interactive. Sydney was created using Poser 7 which is a dynamic three-dimensional rendering and animation software tools produced by Smith Micro Software. Much research and planning had been done before proceeding to the construction of the animation video and other databases. During the development process, many hours had been spent on the manipulating Sydney to sign ASL appropriately and creating the appropriate animation video for the GUI. Due to the large file size of the raw video produced by Poser 7, another application, Auto Gordian Knot, had been used to convert and compress raw AVI video into AVI video with XviD codec. Speech recognition software, Dragon NaturallySpeaking 10, was also successfully implemented to work together with the application. Text to speech was later introduced using Cepstral Diane US English voice and the extraction of the voice was done using the TextToWav application. Through the help of many software application tools, the database of Let’s Talk had been successfully created.