Gesture Language Translation Using Convolutional Neural Network

This study presents a revolutionary gesture language translator that can convert sign language into both text and speech output, which can aid in communication for people who are deaf or mute. This method differs from other existing alternatives in that it can convert gestures into text and voice and the other way around, from spoken words into corresponding gestures. By enabling smooth interaction between the hearing and the non-hearing, this dual-mode system improves communication. The system uses a Convolutional neural network (CNN) for gesture recognition and Flask and Python for back-end operations. Google's text-to-speech (TTS) API generates voice output, while Media-pipe is utilized to detect hand landmarks in real-time. In the opposite procedure, vocal inputs are recorded by speech recognition technology and converted into gestures using a pre-stablished sign language model. This product represents a significant achievement in the field of assistive technologies for people with speech and hearing impairments since it has a distinctive feature: two-way translation capabilities.