SafeStep is a full-stack web application designed to streamline the creation, management, and accessibility of Standard Operating Procedures (SOPs) for the manufacturing industry.
By leveraging advanced AI models, SafeStep allows managers to generate detailed SOPs from scratch (via machine parameters) or by extracting context from existing manuals. It then converts those technical documents into worker-friendly conversational scripts and synthesizes them into highly customized, multilingual audio tracks.
- Context Engine: Initialize SOPs by defining operational parameters (Machine Type, PPE, Worker Expertise) or by uploading existing PDF manuals for AI text extraction.
- 7-Step Drafting Wizard: Step-by-step AI-assisted drafting for standard SOP sections (Purpose, Scope, References, Definitions, Roles, Procedure, Appendices).
- Script Simplifier: Automatically translates dense, technical SOPs into conversational, easy-to-understand scripts designed for floor workers.
- Voice Design Engineering: Integrates with the ElevenLabs Voice Design API to generate dynamic, custom AI voices tailored by age, gender, emotion, and pacing.
- SOP Library: A centralized dashboard to view, download (PDF/Audio), edit, and manage all generated SOPs.
Frontend:
- React 18
- Vite (Build tool & development server)
- Tailwind CSS (Styling)
Backend:
- Node.js & Express
- MongoDB & Mongoose (Database)
- Multer & pdf-parse (File uploading & document text extraction)
Third-Party Services & APIs:
- Google Gemini API (
gemini-flash-latest): Used for PDF summarization, contextual drafting, and script simplification. - ElevenLabs API: Used for text-to-speech and on-the-fly Voice Design generation.
- Cloudflare R2: S3-compatible object storage used to host uploaded manuals, generated
.txtscripts, and.mp3audio files.
Follow these instructions to set up and run the project locally.
Make sure you have the following installed and set up on your machine:
- Node.js (v18 or higher recommended)
- Git
- A MongoDB cluster (e.g., MongoDB Atlas)
- A Cloudflare account with an R2 Bucket created
- API Keys for Google Gemini and ElevenLabs
git clone https://github.com/YourUsername/YourNewRepo.git
cd YourNewRepo- Navigate to the backend directory:
cd backend - Install the backend dependencies:
npm install
- Create a
.envfile in thebackenddirectory and add your secret keys. Do not commit this file!PORT=5000 MONGO_URI=your_mongodb_connection_string GEMINI_API_KEY=your_gemini_api_key ELEVENLABS_API_KEY=your_elevenlabs_api_key # Cloudflare R2 Storage Configurations R2_ACCOUNT_ID=your_cloudflare_account_id R2_ACCESS_KEY_ID=your_r2_access_key R2_SECRET_ACCESS_KEY=your_r2_secret_key R2_BUCKET_NAME=your_r2_bucket_name R2_PUBLIC_URL=your_r2_public_routing_url
- Open a new terminal window/tab and navigate to the frontend directory:
cd frontend - Install the frontend dependencies:
npm install
You will need to run the backend and frontend servers simultaneously in two separate terminal windows.
Start the Backend Server:
cd backend
npm run dev(The backend runs on http://localhost:5000)
Start the Frontend Server:
cd frontend
npm run dev(The frontend runs on http://localhost:5173)
Once both are running, open your browser and navigate to http://localhost:5173 to use SafeStep!
Developed for the modern manufacturing floor.