Patrik Gajdoš

About the Author

I specialize in developing tools that combine artificial intelligence with user protection in the online environment. This platform was created as part of my master's thesis at the Department of Computer Networks (KPS), FEI TUKE. I also work as a show designer for HC Košice, creating visual and lighting productions for game events.

About the Project

The platform combines text conversation analysis, risky behaviour detection and voice-based age estimation.

Its goal is to support cyber-threat prevention and make modern AI technologies accessible to the public.

2

modules (text and voice analysis)

70 000+

data records

4

trained AI models

Project Story

The project started as research on online harassment detection and expanded to voice processing and multimodal evaluation.

It included designing AI models, building API services, UI development and real-world testing.

Used Technologies

TensorFlow, Keras
Transformers (mDeBERTa, RoBERTa, GPT-J)
FastAPI, PHP
MFCC, spectrograms, NLP pipeline
Hugging Face Spaces
SQL, data pipelines

A safe digital environment is the foundation of modern society. Our role is to create tools that help people stay protected.

Thesis Supervision

Supervisor: doc. Ing. Matúš Pleva, PhD.

Consultant: Ing. Eva Kupcová

KPS – FEI TUKE

The Department of Computer Networks (KPS) is a key department at FEI TUKE focusing on education, research and innovation in IT systems.

Visit KPS Visit TUKE Visit FEI

Future Vision

The platform will be expanded with real-time voice analysis, deepfake detection and new digital safety tools.