An Investigation on Speech to Sign Language Translator for Hearing Impaired (SSLT) using Machine Learning Techniques

The language of the hearing impaired people is a visual language. It has to transmit sound patterns in the form of signs. A person’s thoughts are expressed by hand signals and facial expressions. Deaf people normally get struggle by making conversation with normal people; The languages will be different for the various group of deaf people all over the world as well. Our project is to guide the hearing deaf or speech defected persons made communicating with normal persons. It automatically translates the speech in English into Indian sign language. It is the sign-language translating system. For the communication between the normal person and the impaired persons, it could be used as a translator for their natural way of speaking. It’s helpful for people who didn't understand sign language, the sign language gestures are emojis. Our proposed model brings the accuracy improvement by 10% compare to existing models of CNN based classification and SVM_HMM models. FPV and PPV improved by 23.52 % and 37.34 %.