Enhancing Communication: Automated Translation of Amharic Text to Ethiopian Sign Language using Deep Learning Approach

Communication is the day-to-day life of communities; good communication channels are crucial for mutual understanding. the hearing-impaired community uses sign language. With more than 300 sign languages across the world, sign interprets are not always available to translate spoken words into sign language and vice versa. Even though sign language is an excellent platform for communication among hearing impaired persons, it has created a communication barrier between hearing-impaired and non-disabled people. this would limit their communication with others. To address this issue, utilizing researchers have proposed text to sign language translation systems for English and other European languages as a solution. As I consider the success of Deep Learning for Computer Vision tasks and gestural language I observe the advantage it can provide for Enhancing Communication: Automated Translation of Amharic Text to Ethiopian Sign Language. For this research, optimizing our model by tuning our hypermeters. The main aim was to optimize the model, achieve better accuracy, and lower log loss.   The proposed system was created with the help of two key deep learning algorithms: the stochastic nature of deep learning model and a Long Short-Term Memory (LSTM). results can vary even when Train the sequence-to-sequence model end-to-end, optimizing the model parameters to minimize the loss between the predicted Ethiopia sign language output and the ground truth sign language sequences in the dataset. Before feeding the obtained dataset to the deep learning models, data preprocessing activities such as cleaning and videoto-image frame segmentation were conducted. The LSTM was used to extract sequence information from a sequence of image frames of a specific sign language, while to the stochastic nature of the deep learning model was used to extract features from single-frame images. Due to the stochastic nature of deep learning models build a recognition system for Automated Translation of Amharic Text to Ethiopian Sign Language the signs for the Ethiopian Sign Language. The system tested an LSTM model on a signer-independent dataset. I achieved 100% accuracy on the testing dataset on our own dataset. The proposed system can be utilized as a starting point for other researchers to improve upon, based on the outcomes of the experiment. Further high- quality training datasets and highperformance training machines, similar as GPU- enabled computers, can be added to the system to ameliorate it.