VoiceMitra - A Speech to Visual Translation Approach for Disabled

Nowadays, Internet’s expansion has multiplied greatly with a variety of information in all contexts available. Although it has been beneficial for the majority group of people, those with special needs, such as the deaf, have few resources at their disposal. The visual graphics and hand gestures can be used to communicate in such cases. The objective of this research is to develop a system that transforms speech and text input into a sequence of visuals representing sign language. This system basically consists of three modules- 1. Speech to Text, which converts the spoken audio into text format, 2. Language translator, wherein different standard languages can also be given as input to get the output and 3. Sign Language Converter, through which the given audio/text is converted to either hand gestures or related gif. These graphics are understood by them, and so a well understood communication can be established. The language translation modules give an average confidence level of 0.916823557. In addition to these three modules, the feature of saving the history in multiple languages is also added, making this system a user-friendly system useful in future as well.