Neural_Machine_Translation

Implementation of a Machine Translator using a Sequence to Sequence LSTM Network (Encoder-Decoder) with Attention in Tensorflow.

Dataset

We are using the Open Subtitles Dataset which is a collection of documents from OpenSubtitles. So basically the data comprises of movie subtitles in different languages. We have chosen data for English-> German task and French-> English translation task.

Note - Some language data is in a TMX format, which needs to be converted in a useful format before we train our model. I have written a TMX convertor which accomplishes this task, and can be used on any language dataset for creating a machine translator.

Dependencies

1. python 2.7
2. tensorflow 1.0.1
3. numpy 1.13.0
4. scikit-learn 0.18.2
5. matplotlib 1.5.1

Implementation

1. English to German text translation

2. French to English text translation

Results

1. English -> German

1. What' s your name -> Was ist denn Name

2. My name is -> Mein Name ist

3. What are you doing -> Was machst du gemacht

4. I am reading a book -> Ich bin ein Buch Buch

5. How are you -> Wie geht' s

French -> English

1. Quel est ton nom -> What is your name

2. Mon nom est -> My name is

3. Qu'est-ce que tu fais -> You are wrong

4. Oui -> Yeah

5. Non -> No