Deaf-and-mute sign language generation system

摘要

We have developed a system which can recognize speech and generate the corresponding animation-like sign language sequence in real-time.The system is implemented in a popular personal computer. This has three video RAM (VRAM) and a voice recognition board which can recognize only the registered voice of a specific speaker. At present, forty sign language patterns and fifty finger spellings are stored on two floppy disks.Each sign pattern is composed of one to four sub-patterns. That is, if the pattern is composed of one subpattern, it is displayed as a still pattern. If not, it is displayed as a motion pattern. Japanese has fifty Kana characters. Each character has a corresponding finger spelling. Some of them are also displayed as a motion pattern composed of three sub-patterns.When a text is input from a keyboard or voice recognition circuit board (unit), a corresponding pattern sequence composed of a mixture of sign language patterns and finger spelling patterns is generated and displayed on a CRT. The data of each subpattern are read from a floppy disk and transferred to one of the VRAM (640 × 200 × 1 bits). Since the computer has three VRAM, it is possible to read and transfer the data even when a preceding pattern is still being displayed on the CRT. Thus, spoken natural language sentences can be displayed in animation form by switching the VRAM.This system will help communication between deaf-mute and healthy persons. In order to display in high speed, most programs are written in a machine language.