SIGN LANGUAGE TO SPEECH CONVERSION
Abstract
Humans communicate with one another in order to share their ideas, thoughts, and experiences with others around them. For deaf-mute individuals, however, this is not the case. Deaf-mute individuals can communicate thanks to sign language. A deaf-mute individual may communicate without the use of acoustic noises by using sign language. The goal of this project is to create a system for detecting sign language that allows individuals with speech impairments and normal people to communicate, thereby closing the communication gap. Hand gestures are essential because they convey the user's ideas in less time than other motions (arm, face, head, and body). A flex sensor-based gesture recognition module is being created in this project to identify a few phrases and broadcast them via a speaker.
Dumb and deaf individuals make up a large portion of India's population. As a result, the system is developing a glove-based device for converting sign language (ASL) to voice. The fundamental system consists of two parts: recognition of sign language and conversion to text, followed by voice. The sign language glove is made up of basic hand gloves with flex sensors that track the amount of bend in the fingers. The sensors that vary the resistance based on the degree of bend on the sensor are known as flex sensors. The data from the sensors is sent to the CC3200 control unit. The analogue signals from the sensors are digitally translated and compared to the recorded value for sign recognition, after which the text is shown on the android smartphone. In addition, the text output is wirelessly sent to a cellular phone or a PC running test-to-speech software. We are currently working on a simple prototype that will translate basic Alphabets and numbers, which will be expanded for word recognition in the future.
The primary goal is to convert sign language into text or voice. The framework enables speech-impaired people to communicate with the rest of the world via sign language. As a result, the intermediate person, who usually serves as a translator, is no longer required. By offering speech/text output for a sign gesture input, this would provide a user-friendly environment for the user.
