Filipino Deepfake Audio Detection Tool Using Machine Learning, Deep Learning, and Transformer Models
2025
With deepfake audio becoming more prevalent and its potential misuse
in
fraud, misinformation, and identity theft, we developed the first-ever Filipino deepfake audio detection
tool for our thesis project. Our approach combined deep learning models (ResNet-50, VGG-16), transformer
models (Wav2Vec2, HuBERT), and machine learning models (Decision Tree, Extra Trees, Random Forest, Gradient
Boosting). We trained these models on our own Filipino audio dataset to classify deepfake and genuine audio
in the Filipino language. The best-performing models (ResNet-50, VGG-16, HuBERT, and Wav2Vec2) all achieved
an impressive 100% accuracy in classification. Through this project, I gained in-depth knowledge of deep
learning, machine learning models, and transformer-based architectures, and integrating them into web
applications.
Scan and Sort: Mobile Application for Household Waste Classification Using ResNet-50
2024
As part of a deep learning group project, we developed a mobile
app for household waste classification. We trained a ResNet-50 model on our own dataset of 15,000 images
with 15 household waste categories. The model achieved an impressive 93.88% training accuracy and 90.33%
validation accuracy. Through this project, I deepened my understanding of training and fine-tuning deep
learning models, dataset preparation, and model evaluation, as well as deploying them in mobile applications
to address real-world problems like efficient waste management.
iPet Club PH: Online Booking and Management System
2024
Through this project, I took on the role of project leader and lead
developer for an online booking system for iPet Club PH, a pet hotel and salon in BGC, Taguig. This website,
designed using Figma and developed with Bootstrap, HTML, CSS, JavaScript, PHP, and SQL, allowed customers to
make reservations, apply for memberships, and submit inquiries. Additionally, an admin panel was implemented
for managing bookings and applications.
This project not only enhanced my full-stack development skills but also strengthened my project management,
problem-solving, and team coordination abilities. It also helped refine my communication skills, as I
collaborated closely with both clients and team members to ensure the project met all requirements on time.
TranslatePH: Filipino-English Translation Mobile Application
2023
Developed a mobile application that facilitates translation
between Filipino and English using text and voice input. It utilizes Google APIs for Text-to-Speech (TTS)
and Speech-to-Text (STT) for speech recognition and translation. The app also features Firebase
Authentication for secure login and a currency converter with real-time exchange rates to Philippine Pesos
(PHP), making communication more accessible for users looking to learn both English and Filipino.
Momma's Corner PH: E-Commerce Baking Store UI Design
2023
Designed a UI prototype in Figma for Momma’s Corner PH, an online store in
Makati that sells baking tools and equipment. This project focused on creating a visually appealing and
easy-to-use e-commerce experience for both customers and administrators.
Through this project, I gained a better understanding of e-commerce and
UI/UX principles, improving my skills in Figma, design consistency, and prototyping interactive elements to
enhance the overall shopping experience.
NBA Defensive Performance Analysis (2012-2023): Regular Season vs. Playoffs
2025
For a data science project, I analyzed an NBA basketball dataset
to examine defensive metrics such as steals, blocks, and defensive rebounds in the regular season vs.
playoffs. The goal was to identify trends, understand key factors influencing these metrics, and assess
their impact on team performance across different seasons. Despite not being highly knowledgeable about
basketball, I overcame this challenge by conducting thorough research to understand the dataset.
I used Python with libraries such as pandas for data manipulation
and aggregation, seaborn for creating regression plots, and matplotlib for data processing and
visualization. Additionally, I used Canva to create a clear and engaging presentation of my data analysis.
This project enhanced my skills in data analysis, visualization, and presentation, allowing me to extract
insights from large datasets and communicate findings effectively.
Contact Me
Get In Touch
If you have any questions or are interested in working together, feel free to reach out! I’m always open to new opportunities, collaborations, and learning from new projects and experiences. You can contact me via email or through this contact form. I’d love to hear from you!