Anti-Bullying Notification Application Using Voice Recognition via ESP32 and OpenAI
Specifications
- Code
- SA2602
- Department
- DKS
- Technology
- ESP32 · OpenAI
- Theme
- Education & Society · Safety & Security
- Student
- Muhammad Nursyafiq bin Khairul Amree
- Supervisor
- Ja'aris bin Samsudin
Objectives
Main Objective: To develop an automated system that detects potential bullying-related speech and sends timely alerts to teachers. Specific objectives: Establish a stable and reliable Wi-Fi connection between the ESP32 and the internet. Design and fabricate a custom PCB integrating the required hardware components into one unit. Develop an Android notification application that receives real-time alerts and allows teachers to acknowledge incidents.
Scope of project
Record conversation during incidents using a microphone Convert recorded audio into text using whisper AI Analyze transcribed conversations using GPT API Require an active Wi-Fi for cloud AI processing Send instant notifications to teachers via application Design and develop the complete electronic circuit
Project description
The Anti-Bullying System is an AI-powered ESP32 device that continuosly listens for bullying in school secluded areas such as toilet and stairwell. It records the conversation and transcribes the speech in English and Malay using OpenAI Whisper, analyses context with GPT-4o mini, and it instantly alerts teachers through a local alarm and mobile notification, enabling earlier detection and faster intervention.