ConText is a desktop app for secure, local text translation using Ollama LLMs. It supports language detection, text-to-speech, web scraping, text summarization, and YouTube transcript extraction, all processed locally.
| Feature | Description |
|---|---|
| Local translation | Process translations using Ollama models locally |
| Language detection | Automatically identify the language of input text |
| Text-to-speech | Convert text to WAV audio files |
| Web content scraping | Extract text from websites for translation |
| Text summarization | Create concise summaries of longer texts |
| YouTube transcript retrieval | Get transcripts from YouTube videos |
| Multiple language support | Translate between various languages |
| Configurable text chunking | Customize text processing parameters |
- Python 3.8+
- Ollama installed and running
- Clone the repository:
git clone https://github.com/KazKozDev/ConText.git cd ConText- Set up the backend:
cd backend
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
pip install -r requirements.txt- Start Ollama:
ollama serve- Start the backend:
cd backend
source venv/bin/activate # On Windows: venv\Scripts\activate
python app.pyBackend runs on http://localhost:5002.
| Endpoint | Function | Description |
|---|---|---|
| /health | Server status | Check if the server is running properly |
| /translate | Translate text | Convert text between languages |
| /detect-language | Detect language | Identify the language of input text |
| /tts | Text-to-speech | Convert text to audio format |
| /scrape-url | Scrape web content | Extract text from web pages |
| /summarize | Summarize text | Create concise summaries of texts |
| /youtube-transcript | YouTube transcript | Extract transcripts from YouTube videos |
Translate English to Spanish:
curl -X POST http://localhost:5002/translate \
-H "Content-Type: application/json" \
-d '{"text": "Hello, world!", "source_lang": "en", "target_lang": "es"}'Response:
{"translated_text": "¡Hola, mundo!"}If you like this project, please give it a star ⭐
For questions, feedback, or support, reach out to: