A professional Streamlit-based call transcription system that converts audio in any language to English text with AI-powered analysis.
pip install -r requirements.txt
Create a .env file:
SUPABASE_URL=your_supabase_url
SUPABASE_KEY=your_supabase_key
python -m streamlit run app.py
Open http://localhost:8501 in your browser.
reception-agent/
├── app.py # Main Streamlit application
├── auto_processor.py # Background folder watcher (optional)
├── requirements.txt # Python dependencies
├── .env # Environment variables (create this)
├── .gitignore # Git ignore file
├── .streamlit/
│ └── config.toml # Streamlit configuration
├── README.md # This file
├── DEPLOYMENT.md # Deployment instructions
└── GITHUB_UPLOAD_STEPS.md # GitHub upload guide
Set DEFAULT_LANGUAGE in .env to match your primary language:
| Language | Code |
|---|---|
| English (US) | en-US |
| Hindi | hi-IN |
| Kannada | kn-IN |
| Tamil | ta-IN |
| Telugu | te-IN |
| Malayalam | ml-IN |
| Spanish | es-ES |
| French | fr-FR |
| German | de-DE |
| Chinese | zh-CN |
Run the background processor to automatically transcribe files placed in a folder:
python auto_processor.py
This monitors C:/CallRecordings for new audio files and processes them automatically.
call_records with these columns:
id (int8, primary key)timestamp (timestamptz)filename (text)transcribed_text (text)analysis (text)language (text)See DEPLOYMENT.md for instructions on deploying to Streamlit Cloud.
pydub - Audio processing