Russian to English Translator

This Python script translates Russian text files to English using the MarianMT model. It supports various optional features for improving translation quality.

Features

Translates Russian text files to English
Automatically detects input file encoding
Handles large texts by breaking them into paragraphs and chunks
Optional back-translation quality check
Optional ensemble translation using multiple models
Optional confidence score checking
Customizable output file naming
Docker support for easy setup and execution

Requirements

If running natively:

Python 3.6 or higher
PyTorch
Transformers library
NLTK
chardet

If using Docker:

Docker

Installation and Usage

Option 1: Using Docker (Recommended)

Ensure you have Docker installed on your system.

Clone this repository:

git clone https://github.com/yourusername/russian-to-english-translator.git
cd russian-to-english-translator

Build the Docker image:
```
docker build -t russian-translator .
```

Run the translator using Docker:

docker run -v $(pwd):/app russian-translator /app/input_file.txt -o /app/output_file.txt

Replace /app/input_file.txt with the path to your input file inside the container, and /app/output_file.txt with your desired output path.

To use optional improvements:

docker run -v $(pwd):/app russian-translator /app/input_file.txt -o /app/output_file.txt --use-back-translation --use-ensemble --use-confidence-check

Option 2: Native Installation

Ensure you have Python 3.6+ installed on your system.

Clone this repository:

git clone https://github.com/yourusername/russian-to-english-translator.git
cd russian-to-english-translator

Install the required packages:
```
pip install -r requirements.txt
```

Run the translator:

python translate_with_dl.py path/to/input_file.txt

Usage (Native Installation)

Basic Translation

To translate a Russian text file to English:

python translate_with_dl.py path/to/input_file.txt

This will create an output file named output_english_text.txt in the current directory.

Optional Improvements

You can enable various improvements to potentially enhance translation quality:

Back-translation Quality Check:

python translate_with_dl.py path/to/input_file.txt --use-back-translation

Ensemble Translation (uses multiple models):

python translate_with_dl.py path/to/input_file.txt --use-ensemble

Confidence Score Checking:

python translate_with_dl.py path/to/input_file.txt --use-confidence-check

You can combine these options as needed:

python translate_with_dl.py path/to/input_file.txt --use-back-translation --use-ensemble --use-confidence-check

Specifying Output File

To specify a custom name or path for the output file:

python translate_with_dl.py path/to/input_file.txt -o path/to/output_file.txt

Note

The first run may take longer as it downloads the necessary models and NLTK data.
Enabling optional improvements will increase translation time and resource usage.

genaforvena/translator