Logo Lanfrica

bjeptum/nochai-ussd-ai

Domaine:

natural language processingpeace and security

Type de record:

softwaretools
Créateur:
bje
Hôte:
An intelligent Swahili-first system that helps everyday Kenyans refuse and report micro-corruption via USSD and WhatsApp. # NoChai — An Intelligent Bribe Refusal Coach for Kenya An intelligent Swahili-first system that helps everyday Kenyans refuse and report micro-corruption via USSD and WhatsApp. **Theme:** Governance & Public Policy | Combating Corruption ## Why This Matters - 70% of Kenyans pay bribes monthly (EACC) - Gen-Z protests fueled by graft in public services - Rural citizens lose fertilizer subsidies to "kitu kidogo" NoChai turns victims into empowered whistleblowers using **voice + USSD + Swahili AI** ## Key Features - Dial `*384*xxxx#` on any feature phone and report in Sheng/Swahili - AI generates **real-time refusal scripts** - Anonymous reports logged to public **Corruption Heatmap** - Works on WhatsApp with rich responses - Built for Kenya with multilingual support (Swahili, Sheng, English) ## Tech Stack - **Backend:** Python 3.8+ + Flask - **USSD:** Africa's Talking API - **WhatsApp:** Twilio Sandbox - **AI Model:** BLOOM-560M (multilingual language model) - **Framework:** LangChain + HuggingFace Transformers - **Database:** SQLite - **Dashboard:** Streamlit + Folium - **Tunneling:** Ngrok ##Set up & Installation ### Prerequisites - Python 3.8 or higher - Git - Virtual environment tool (venv) - Ngrok account (free tier) - Hugging Face account (for model access) ### Step 1: Clone the Repository ```bash git clone github.com cd nochai-ussd-ai ``` ### Step 2: Create Virtual Environment ```bash # Create virtual environment python3 -m venv venv # Activate it source venv/bin/activate # On Linux/Mac # OR venv\Scripts\activate # On Windows ``` ### Step 3: Install Dependencies ```bash # Upgrade pip first pip install --upgrade pip # Install required packages pip install -r requirements.txt # Download spaCy English model python3 -m spacy download en_core_web_sm ``` ### Step 4: Configure Environment Variables Create a `.env` file in the project root: ```bash # .env HUGGINGFACE_API_TOKEN=your_huggingface_token_here ASSEMBLYAI_AP …

Languages