Skip to content

Implementation of a Machine Translator using a Sequence to Sequence LSTM Network with Attention

Notifications You must be signed in to change notification settings

darshanbagul/Neural_Machine_Translation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

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

About

Implementation of a Machine Translator using a Sequence to Sequence LSTM Network with Attention

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published