Machine Translation: Spanish-to-English Translation System using RNNs
Hrithick Gokul Y, N Dinesh, R Hariharan · International Journal for Research in Applied Science and Engineering Technology · 2022
Abstract: Machine translation has undergone a revolution in recent years. Older translation systems developed by linguists utilizing decades of statistical study have been superseded by new systems developed using deep learning. The core idea of a translation system will be a neural network that takes in input and generates the translation of a given sentence. This paper discusses the model which uses intermediate numerical encoding to learn the mapping from one human language to another. Two neural networks connected end-to-end provide the basis of a neural translation system. The first neural network encodes sequences of words in Spanish into an array of numbers with the same meaning. The second neural network learns to translate the numbers back into a string of words that all have the same meaning in English.