| # This line tells the server that this is a shell script | |
| # Launch the FastAPI backend API in the background. | |
| # The '&' symbol at the end is crucial. It tells the server to run this command | |
| # and immediately move on to the next one without waiting for it to finish. | |
| echo "Starting backend server..." | |
| uvicorn v2_multilingual_api.backend.main:app --host 0.0.0.0 --port 8000 & | |
| # Give the backend a moment to initialize the AI models. | |
| # Loading models can take time. This 'sleep' command pauses for 20 seconds | |
| # to ensure the backend is ready before the frontend starts trying to talk to it. | |
| echo "Waiting for backend to initialize..." | |
| sleep 20 | |
| # Launch the Streamlit frontend. | |
| # This is the final command, so it runs in the foreground. | |
| # --server.port 7860 is the required port for Streamlit apps on Hugging Face. | |
| # --server.address 0.0.0.0 makes it accessible. | |
| echo "Starting frontend server..." | |
| streamlit run v2_multilingual_api.frontend/app.py --server.port 7860 --server.address 0.0.0.0 | |