https://github.com/enricopesce/translator
A simple tool to create text translation service using OCI GenAI LLM service
https://github.com/enricopesce/translator
Last synced: 9 months ago
JSON representation
A simple tool to create text translation service using OCI GenAI LLM service
- Host: GitHub
- URL: https://github.com/enricopesce/translator
- Owner: enricopesce
- Created: 2024-11-27T09:54:57.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2024-12-05T10:43:42.000Z (about 1 year ago)
- Last Synced: 2025-01-19T05:54:25.018Z (11 months ago)
- Language: Python
- Homepage:
- Size: 263 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Translator
## Introduction
Translator demonstrates the power of using [Oracle Cloud Infrastructure's Generative AI services](https://www.oracle.com/artificial-intelligence/generative-ai/generative-ai-service/) to build practical language based applications.
This project showcases how OCI Generative AI services can translate text with a modern LLM that understands context and nuances and how to integrate it into a modern web application.
### Why OCI Generative AI?
Oracle Cloud Infrastructure (OCI) Generative AI offers several compelling advantages for translation services:
- **Enterprise-Ready**: Built on OCI's robust infrastructure, ensuring high availability and scalability
- **Cost-Effective**: Pay-as-you-go pricing model with competitive rates for API calls
- **Security-First**: Enterprise-grade security with built-in governance and compliance features
- **High Performance**: Low-latency responses suitable for real-time translation needs
- **Multilingual Support**: Comprehensive support for major world languages with high accuracy
- **Context Awareness**: Advanced language models that understand context and nuances
- **Integration Friendly**: Easy integration with existing cloud infrastructure and applications
# Supported Languages
| Language | ISO Code |
| ---------- | -------- |
| Arabic | ar |
| Chinese | zh |
| English | en |
| French | fr |
| German | de |
| Italian | it |
| Japanese | ja |
| Korean | ko |
| Portuguese | pt |
| Spanish | es |
## Prerequisites
- Python 3.8 or higher
- Oracle Cloud Infrastructure account with access to Generative AI service
- OCI credentials configured
## Installation
1. Clone the repository:
```bash
git clone https://github.com/enricopesce/translator
cd translator
```
2. Install required packages:
```bash
pip install -r requirements.txt
```
## Configuration
1. Create a `.env` file in the project root with your OCI credentials (env.example provided):
```env
OCI_MODEL_ID=your_model_id
OCI_SERVICE_ENDPOINT=your_service_endpoint
OCI_COMPARTMENT_ID=your_compartment_id
```
## Usage
1. Start the server:
```bash
python main.py
```
2. The API will be available at `http://localhost:8000`
### API Endpoint
#### POST /translate
Translates text between supported languages.
Request body:
```json
{
"text": "Hello world",
"source_language": "en",
"target_language": "es"
}
```
Response:
```json
{
"translated_text": "Hola mundo",
"source_language": "en",
"target_language": "es"
}
```
### Example Usage with cURL
```bash
curl -X POST "http://localhost:8000/translate" \
-H "Content-Type: application/json" \
-d '{
"text": "Hello world",
"source_language": "en",
"target_language": "es"
}'
```
## Error Handling
The API returns appropriate HTTP status codes and error messages:
- 400: Bad Request (invalid language code/name or impossible translation)
- 500: Internal Server Error (unexpected errors)
Error response format:
```json
{
"detail": {
"message": "Error description",
"source_language": "Source language",
"target_language": "Target language"
}
}
```
## Testing by script
### Basic Usage
```bash
# Run with default settings (localhost:8000)
python test.py
# Run with custom API URL
python test.py --url http://your-api-url
```
### Sample Testing Mode
Running without parameters will execute tests using predefined sample texts in all supported languages:
```bash
python test.py
```
### Single Translation Mode
For testing a specific text with defined source and target languages:
```bash
python test.py --text "Your text here" --from en --to es
```
Parameters:
- `--text`: The text you want to translate
- `--from`: Source language code
- `--to`: Target language code
Example:
```bash
# Translate from English to Spanish
python test.py --text "Hello world" --from en --to es
# Translate from French to German
python test.py --text "Bonjour le monde" --from fr --to de
```
## Output
```bash
python test.py --text "Il più grande nemico della conoscenza non è l'ignoranza, ma l'illusione della conoscenza" --from it --to zh
╭─────────────────────────────────────────────── Translation Testing ────────────────────────────────────────────────╮
│ Starting single translation test │
│ API URL: http://localhost:8000 │
│ From: Italian │
│ To: Chinese │
│ Text: Il più grande nemico della conoscenza non è l'ignoranza, ma l'illusione della conoscenza │
│ Timestamp: 2024-12-05 10:42:45 │
╰────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
Translation #2
From: Italian
To: Chinese
Original: Il più grande nemico della conoscenza non è l'ignoranza, ma l'illusione della conoscenza
Translated: 最大的敌人不是无知，而是知识的错觉。
Time: 0.811s
Characters: 106
--------------------------------------------------------------------------------
================================================================================
Translation Summary
Total translations: 1
Failed translations: 0
Total time: 0.81s
Average time per translation: 0.811s
Total characters processed: 106
Average characters per translation: 106.0
================================================================================
```