Abstract
Building an Effective MT System for English-Hindi Using RNN's
Highlights
Deep learning is a rapidly advancing approach to machine learning and has shown promising performance when applied to a variety of tasks like image recognition, speech processing, natural language processing, cognitive modelling and so on
We demonstrate that the best performance for English > Hindi MT is generally obtained using Bi-directional Long Short Term Memory Units (LSTMs) with attention mechanism and in some cases with Gated Recurrent Units (GRUs) with attention mechanism
The input n-gram is projected into an embedding space for each word and passes to big output layer. This novel idea was used by several researchers who tried to integrate it with Machine Translation systems ((Auli et al, 2013) and (Cho et al, 2014)). (Sutskever et al, 2014) was a breakthrough for Machine Translation, introducing the ”seq2seq” (Sequence to sequence) model which was the first model based completely on neural networks and achieving accuracy comparable to the State-of-the-Art SMT systems
Summary
Deep learning is a rapidly advancing approach to machine learning and has shown promising performance when applied to a variety of tasks like image recognition, speech processing, natural language processing, cognitive modelling and so on. This paper demonstrates the application of deep learning for Machine Translation of English ! The application of deep neural networks to Machine Translation has been demonstrated by (Kalchbrenner and Blunsom, 2013; Sutskever et al, 2014; Cho et al, 2014; Bahdanau et al, 2014) and it has shown promising results for various language pairs. We experiment with different deep learning architectures. These include Gated Recurrent Units (GRUs), Long Short Term Memory Units (LSTMs) and addition of attention mechanism to each of these architectures. We demonstrate that the best performance for English > Hindi MT is generally obtained using Bi-directional LSTMs with attention mechanism and in some cases with GRUs with attention mechanism.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
