Theopilus Bayu Sasongko

Theopilus Bayu Sasongko

Auto-generated profile, based on publicly available information

Theopilus Bayu Sasongko

Universitas AMIKOM Yogyakarta

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 January 2025 in IEEE Access
Verified
Journal Article
Open access
Setyanto AriefSetyanto Arief
Setyanto Arief
TSTheopilus Bayu Sasongko
Theopilus Bayu Sasongko
MFMuhammad Ainul Fikri
Muhammad Ainul Fikri
DADhani Ariatmanto
Dhani Ariatmanto
IAI. Made Artha Agastya
I. Made Artha Agastya
In kee KimIn kee Kim
In kee Kim
Edge computing, a distributed computing paradigm that places small yet capable computing devices near data sources and IoT sensors, is gaining widespread adoption in various real-world applications, such as real-time intelligent drones, autonomous vehicles, and robotics. Object detection (OD) is an essential task in computer vision. Although state-of-the-art deep learning-based OD methods achieve high detection rates, their large model size and high computational demands often hinder deployment on resource-constrained edge devices. Given their limited memory and computational power, edge devices like the Jetson Nano (J. Nano), Jetson Orin Nano (Orin Nano), and Raspberry Pi 4B (Raspi4B) require model optimization and compression techniques in order to deploy large OD models such as YOLO. YOLOv4 is a widely used OD model with a backbone for image feature extraction and a prediction layer. Originally, YOLOv4 was designed to use CSPDarkNet53 as its backbone, which requires significant computational power. In this paper, we propose replacing its backbone with a smaller model, such as MobileNetV2 and RepViT. In order to ensure the strong backbone performance, we perform knowledge distillation (KD), using CSPDarknet53 as the teacher and the smaller model as the student.We compare various KD algorithms to identify the technique that produces a smaller model with the modest accuracy drop. According to our experiments, Contrastive Representation Distillation (CRD) yields MobileNetV2 and RepViT with an acceptable accuracy drop. We consider both accuracy drop and model size to choose either MobileNetV2 or RepViT model to replace CSPDarknet53 in the modified YOLOv4 named M-YOLO-CRD and RV-YOLO-CRD. Our evaluation results demonstrate that RV-YOLO-CRD reduces 30% of model size and achieves better mean average precision (mAP) than M-YOLO-CRD. Our experiments show that the M-YOLO-CRD significantly reduces model size (from 245.5 MB to 35.76 MB) and inference time (6× faster on CPU, 4× faster on J. Nano, and 2.5× faster on Orin Nano. While the precision decreased slightly (less than 4%) , the model still performs well on edge devices. The M-YOLO-CRD achieved latency per frame at around 37 ms on Orin Nano, 168 ms on J. Nano, and 1310 ms on Raspi4B.

31 July 2026 in Intechno Journal (Information Technology Journal)
Verified
Journal Article
Open access
PDF file
MRM. Nasyid Yunitian Rizal
M. Nasyid Yunitian Rizal
TSTheopilus Bayu Sasongko
Theopilus Bayu Sasongko
AHArifiyanto Hadinegoro
Arifiyanto Hadinegoro
KYKumara Ari Yuana
Kumara Ari Yuana
WAWahid Miftahul Ashari
Wahid Miftahul Ashari
Voice-based age and gender recognition plays an important role in biometric authentication, personalized human–computer interaction, and digital forensic applications. However, existing single deep learning architectures often struggle to simultaneously capture local acoustic patterns, temporal dependencies, and long-range contextual information from speech signals. This study proposes a hybrid CNN–BiLSTM–Transformer framework to improve the accuracy and robustness of multi-class age and gender classification. The proposed approach employs Mel-spectrogram representations generated from the Mozilla Common Voice dataset, followed by audio standardization and feature extraction. A Convolutional Neural Network (CNN) enhanced with Squeeze-and-Excitation blocks extracts discriminative spectral features, a Bidirectional Long Short-Term Memory (Bi-LSTM) network models bidirectional temporal dependencies, and a Transformer Encoder captures global contextual relationships through a multi-head self-attention mechanism. The model was evaluated on 40,392 speech samples across 12 age–gender categories. Experimental results achieved an overall classification accuracy of 91%, outperforming standalone CNN and Bi-LSTM models, which obtained accuracies of 84% and 74%, respectively. In addition, the proposed model demonstrated balanced performance with macro-average precision, recall, and F1-scores of 0.91, 0.92, and 0.91. The novelty of this research lies in the integration of complementary spatial, temporal, and global attention mechanisms within a unified architecture for large-scale multi-class voice-based demographic classification, providing an effective and scalable solution for intelligent biometric and speech analysis systems.