Controller-Side Reliability of Edge-Deployed Thai Speech and Gesture Smart-Load Control on Raspberry Pi 5

Authors

Neres Supalugsanasuksakorn
Computer Technology Department, Khon Kaen Technical College, Khon Kaen, Thailand
Sakron Taeanonjiew
Electronic Technology Program, Khon Kaen Technical College, Northeastern Vocational Education Institute 3, Khon Kaen, Thailand
Apinun Khonmanee
Electronic Technology Program, Khon Kaen Technical College, Northeastern Vocational Education Institute 3, Khon Kaen, Thailand
Saweth Hongprasit
Department of Computer Engineering, Faculty of Engineering, Rajamangala University of Technology Isan, Khon Kaen Campus, Khon Kaen, Thailand
Wachirapunya Punyawong
Electronic Technology Program, Khon Kaen Technical College, Northeastern Vocational Education Institute 3, Khon Kaen, Thailand.
DOI: https://doi.org/10.52549/ijeei.v14i3.8104
Download PDF
Published: 30-09-2026 • Issue: Vol. 14 No. 3: September 2026 • Section: Regular paper

Abstract

This paper presented an edge-deployed smart-load control system on Raspberry Pi 5 using independent Thai speech and vision-based finger-gesture channels mapped to a common ten-command GPIO actuation layer. Rather than proposing a new recognition algorithm, the study evaluated controller-side execution reliability using GPIO-confirmed binary outcomes and supplemented the human-trial evaluation with system-level engineering profiling. Structured trials examined interaction distance and background speech, yielding 90.0% execution success for both voice (54/60) and gesture (63/70), with different condition-specific performance patterns. A separate 10.5-min integrated profiling run measured computational behavior on the Raspberry Pi 5. Median processing latency was 46.90 ms for Vosk audio blocks and 72.15 ms for MediaPipe hand processing, while median GPIO software-update latency was 0.0688 ms. Median system CPU utilization was 32.30%, process memory was 427.55 MB, and CPU temperature was 83.70 °C; non-zero throttling-status values were observed during sustained workload. These results characterized both execution reliability and resource constraints for non-critical smart-load control on Raspberry Pi-class edge hardware.
Keywords: Embedded systems; Edge AI; Gesture recognition; Reliability evaluation; Dual-modality interaction

References

[1] C. Kim, D. Gowda, D. Lee, J. Kim, A. Kumar, S. Kim, A. Garg, and C. Han, “A review of on-device fully neural end-to-end automatic speech recognition algorithms,” arXiv preprint arXiv:2012.07974, 2021, doi: https://doi.org/10.48550/arXiv.2012.07974.

[2] X. Wang, Z. Tang, J. Guo, T. Meng, C. Wang, T. Wang, and W. Jia, “Empowering Edge Intelligence: A Comprehensive Survey on On-Device AI Models,” ACM Computing Surveys, vol. 57, no. 9, Art. no. 228, 2025, doi: https://doi.org/10.1145/3724420.

[3] W. Shi, J. Cao, Q. Zhang, Y. Li, and L. Xu, “Edge computing: Vision and challenges,” IEEE Internet of Things Journal, vol. 3, pp. 637–646, 2016, doi: https://doi.org/10.1109/JIOT.2016.2579198.

[4] C. Toma, M. Popa, B. Iancu, M. Doinea, and A. Pascu, “Edge Machine Learning for the Automated Decision and Visual Computing of the Robots, IoT Embedded Devices or UAV-Drones,” Electronics, vol. 11, no. 21, Art. no. 3507, 2022, doi: https://doi.org/10.3390/electronics11213507.

[5] C. Irugalbandara et al., “A secure and smart home automation system with speech recognition and power measurement capabilities,” Sensors, vol. 23, no. 13, Art. no. 5784, 2023, doi: https://doi.org/10.3390/s23135784.

[6] L. Lazzaroni, F. Bellotti, and R. Berta, “An embedded end-to-end voice assistant,” Engineering Applications of Artificial Intelligence, vol. 136, Art. no. 108998, 2024, doi: https://doi.org/10.1016/j.engappai.2024.108998.

[7] T. Hobbs and A. Ali, “Smart Home Control Using Real-Time Hand Gesture Recognition and Artificial Intelligence on Raspberry Pi 5,” Electronics, vol. 14, no. 20, Art. no. 3976, 2025, doi: https://doi.org/10.3390/electronics14203976.

[8] S. Kaewwichai, K. Thangthai, P. Tipakorn, and W. Limprasert, “Development and Evaluation of a Thai Automatic Speech Recognition Model Using the Conformer Model,” Science & Technology Asia, vol. 30, no. 3, pp. 60–69, 2025, doi: https://doi.org/10.14456/scitechasia.2025.47.

[9] M. Z. Iqbal and A. G. Campbell, “From luxury to necessity: Progress of touchless interaction technology,” Technology in Society, vol. 67, Art. no. 101796, 2021. doi: https://doi.org/10.1016/j.techsoc.2021.101796

[10] E. Dritsas, M. Trigka, C. Troussas, and P. Mylonas, “Multimodal interaction, interfaces, and communication: A survey,” Multimodal Technologies and Interaction, vol. 9, no. 1, Art. no. 6, 2025, doi: https://doi.org/10.3390/mti9010006.

[11] Y. D. Du, F. Li, and X. Huang, “Research on multimodal human-robot interaction based on speech and gesture,” Computers & Electrical Engineering, vol. 72, pp. 443–454, 2018, doi: https://doi.org/10.1016/j.compeleceng.2018.09.014.

[12] W. Kaczmarek, J. Panasiuk, S. Borys, and P. Banach, “Industrial robot control by means of gestures and voice commands in off-line and on-line mode,” Sensors, vol. 20, no. 21, Art. no. 6358, 2020, doi: https://doi.org/10.3390/s20216358.

[13] P. Huang, I. Ullah, X. Wei, T. A. Ahanger, N. Hassan, and Z. H. Shah, “Towards Energy-Efficient and Low-Latency Voice-Controlled Smart Homes: A Proposal for Offline Speech Recognition and IoT Integration,” arXiv preprint arXiv:2506.07494, 2025, doi: https://doi.org/10.48550/arXiv.2506.07494.

[14] A. P. Rocha, M. Ketsmur, N. Almeida, and A. Teixeira, “An Accessible Smart Home Based on Integrated Multimodal Interaction,” Sensors, vol. 21, no. 16, Art. no. 5464, 2021, doi: https://doi.org/10.3390/s21165464.

[15] N. Malamas, K. Panayiotou, A. Karabatea, E. Tsardoulias, et al., “A multilayer architecture towards the development and distribution of multimodal interface applications on the edge,” Sensors, vol. 24, no. 16, Art. no. 5199, 2024, doi: https://doi.org/10.3390/s24165199.

[16] G. Diraco, G. Rescio, P. Siciliano, and A. Leone, “Review on human action recognition in smart living: Sensing technology, multimodality, real-time processing, interoperability, and resource-constrained deployment,” Sensors, vol. 23, no. 11, Art. no. 5281, 2023, doi: https://doi.org/10.3390/s23115281.

[17] M. M. H. Shuvo, S. K. Islam, J. Cheng, and B. I. Morshed, “Efficient acceleration of deep learning inference on resource-constrained edge devices: A review,” Proceedings of the IEEE, vol. 111, no. 1, pp. 42–91, 2023, doi: https://doi.org/10.1109/JPROC.2022.3226481.

[18] P. Mittal, “A comprehensive survey of deep learning-based lightweight object detection models for edge devices,” Artificial Intelligence Review, vol. 57, Art. no. 242, 2024, doi: https://doi.org/10.1007/s10462-024-10877-1.

[19] S. Heydari and Q. H. Mahmoud, “Tiny Machine Learning and On-Device Inference: A Survey of Applications, Challenges, and Future Directions,” Sensors, vol. 25, no. 10, Art. no. 3191, 2025, doi: https://doi.org/10.3390/s25103191.

[20] S. Di Leo, L. De Cicco, and S. Mascolo, “Real-Time Speech-to-Text on Edge: A Prototype System for Ultra-Low Latency Communication with AI-Powered NLP,” Information, vol. 16, no. 8, Art. no. 685, 2025, doi: https://doi.org/10.3390/info16080685.

[21] L. A. Pazmiño Ortiz, I. F. Maldonado Soliz, and V. K. Guevara Balarezo, “Advancing TinyML in IoT: A Holistic System-Level Perspective for Resource-Constrained AI,” Future Internet, vol. 17, no. 6, Art. no. 257, 2025, doi: https://doi.org/10.3390/fi17060257.

[22] A. González-Docasal, J. Alonso, J. Olaizola, M. Mendicute, M. P. Franco, A. del Pozo, D. Aguinaga, A. Álvarez, and E. Lleida, “Design and Evaluation of a Voice-Controlled Elevator System to Improve the Safety and Accessibility,” IEEE Open Journal of the Industrial Electronics Society, vol. 5, pp. 1239–1250, 2024, doi: https://doi.org/10.1109/OJIES.2024.3483552.

[23] A. El mrabet, A. Tber, M. Benaly, L. Hlou, and R. El gouri, “Unlocking Doors: A TinyML-Based Approach for Real-Time Face Mask Detection in Door Lock Systems,” Indonesian Journal of Electrical Engineering and Informatics, vol. 11, no. 2, pp. 515–527, 2023, doi: https://doi.org/10.52549/ijeei.v11i2.4756.

[24] A. Pervaiz, F. Hussain, H. Israr, M. A. Tahir, F. R. Raja, N. K. Baloch, F. Ishmanov, and Y. B. Zikria, “Incorporating Noise Robustness in Speech Command Recognition by Noise Augmentation of Training Data,” Sensors, vol. 20, no. 8, Art. no. 2326, 2020, doi: https://doi.org/10.3390/s20082326.

[25] D. Povey et al., “The Kaldi speech recognition toolkit,” in Proc. IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), 2011. https://infoscience.epfl.ch/entities/publication/4a69ed57-34b0-41e8-aac7-5b4442e9cf0f/files

[26] Alphacephei, “Vosk offline speech recognition API.” Available: https://alphacephei.com/vosk/ (accessed Jan. 30, 2026).

[27] C. Lugaresi, J. Tang, H. Nash, C. McClanahan, E. Uboweja, M. Hays, F. Zhang, C.-L. Chang, M. G. Yong, J. Lee, W.-T. Chang, W. Hua, M. Georg, and M. Grundmann, “MediaPipe: A framework for building perception pipelines,” arXiv preprint arXiv:1906.08172, 2019, doi: https://doi.org/10.48550/arXiv.1906.08172.

[28] F. Zhang, V. Bazarevsky, A. Vakunov, A. Tkachenka, G. Sung, C.-L. Chang, and M. Grundmann, “MediaPipe Hands: On-device real-time hand tracking,” arXiv preprint arXiv:2006.10214, 2020, doi: https://doi.org/10.48550/arXiv.2006.10214.

[29] E. B. Wilson, “Probable inference, the law of succession, and statistical inference,” Journal of the American Statistical Association, vol. 22, no. 158, pp. 209–212, 1927, doi: https://doi.org/10.1080/01621459.1927.10502953.

[30] D. C. Montgomery, Design and Analysis of Experiments, 9th ed. Hoboken, NJ, USA: John Wiley & Sons, 2017.

Author Biographies

Neres Supalugsanasuksakorn
Senior lecturer in the Computer Technology Department at Khon Kaen Technical College, Thailand. His research interests include the Internet of Things, automatic control systems, programmable logic controller systems, and embedded systems.
Sakron Taeanonjiew
Senior lecturer in Electronic Technology at Khon Kaen Technical College, Northeastern Vocational Education Institute 3, Thailand. His research interests include automatic control systems, electronic circuits and signals, embedded systems, and artificial intelligence.
Apinun Khonmanee
Senior lecturer in the Electronic Technology Program at Khon Kaen Technical College, Northeastern Vocational Education Institute 3, Thailand. His research interests include electronic circuit systems, electrical signaling, and automated control.
Saweth Hongprasit
Lecturer in the Department of Computer Engineering, Faculty of Engineering, Rajamangala University of Technology Isan, Khon Kaen Campus, Thailand. His research interests include CMOS circuit design, analog signal processing, active filters, and oscillators.
Wachirapunya Punyawong
Assistant Professor in the Electronic Technology Program at Khon Kaen Technical College, Northeastern Vocational Education Institute 3, Thailand. His research interests include automatic control systems, remote control systems, embedded systems, and artificial intelligence.

How to Cite

[1]
N. Supalugsanasuksakorn, S. Taeanonjiew, A. Khonmanee, S. Hongprasit, and W. Punyawong, “Controller-Side Reliability of Edge-Deployed Thai Speech and Gesture Smart-Load Control on Raspberry Pi 5”, IJEEI, vol. 14, no. 3, pp. 884–894, Sep. 2026, doi: 10.52549/ijeei.v14i3.8104.