A MULTIFUNCTIONAL SMARTСLOCK FOR VOICE INTERACTION AND ADAPTIVE TASK SCHEDULING

Dmytro Kozliuk, Halyna Klym

Abstract


Background. Current smart devices are often limited to separate functions such as timekeeping, environmental sensing, or voice assistance. This fragmentation hinders a unified solution for productivity in modern workspaces, where indoor conditions and time management are key. Cloud systems face latency and connectivity issues, while local ones are limited by hardware. This study presents a multifunctional smart clock combining environmental monitoring, voice interaction, and task scheduling to enhance comfort, focus, and efficiency.

Methods. The system uses an Edge–Cloud architecture: the ESP32-S3 edge runs latency-critical functions under FreeRTOS with an OOP design. Audio from a MEMS microphone (I²S) is windowed and converted via short-time FFT to log-mel spectrograms; a quantized CNN (TFLM) performs on-device keyword spotting for wake-word detection. After wake-word detection, commands are sent to Wit.ai for ASR/NLU. Audio output is driven by a Class-D amplifier and speaker. Environmental sensing covers temperature, humidity, illuminance, and CO₂ (NDIR), with filtered readings shown in an event-driven LVGL touch GUI and periodically uploaded for analysis to Firebase.

Results and Discussion. The CNN wake-word detector achieved ~90% activation accuracy in quiet-to-moderate office noise with FAR <1 trigger/hour at ~10% FRR; median detection latency remained <200 ms after sufficient context accumulation. Under RTT ≤100 ms, cloud ASR/NLU yielded end-to-end wake→intent latency ≈1–1.5 s. Concurrent environmental monitoring at a 2-s cadence did not perturb the audio pipeline, GUI sustained a 25 Hz refresh rate, and after WWD the system opened a bounded 4-s command window for user utterances. Wi-Fi provisioning via an embedded web server and hourly cloud uploads produced coherent. Threshold-driven voice/visual prompts increased awareness of indoor conditions, while integrated Pomodoro cycles supported sustained focus without auxiliary tools.

Conclusion. The proposed platform integrates voice assistance, time management, and microclimate monitoring in an affordable edge device. Its hybrid speech design balances latency and flexibility, while FreeRTOS ensures reliable multitasking across sensing, GUI, networking, and audio subsystems.

Keywords: Edge computing, RTOS, NDIR, Pomodoro, embedded voice interaction, microclimate monitoring.


Full Text:

PDF

References


  1. Shafik, W. (2024). Smart devices and Internet of Things for sustainable energy. In Advanced Technology for Smart Environment and Energy (pp. 67-93). Cham: Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-50871-4_5
  2. Fakhruldeen, H. F., Saadh, M. J., Khan, S., Salim, N. A., Jhamat, N., & Mustafa, G. (2025). Enhancing smart home device identification in WiFi environments for futuristic smart networks-based IoT. International Journal of Data Science and Analytics, 19(4), 645-658. https://doi.org/10.1007/s41060-023-00489-3
  3. Yedilkhan, D., & Smakova, S. (2024). Machine Learning Approaches for Smart Home Device Recognition from Network Traffic. Procedia Computer Science, 231, 709-714. https://doi.org/10.1016/j.procs.2023.12.157
  4. Masoumian Hosseini, M., Masoumian Hosseini, S. T., Qayumi, K., Hosseinzadeh, S., & Sajadi Tabar, S. S. (2023). Smartwatches in healthcare medicine: assistance and monitoring; a scoping review. BMC Medical Informatics and Decision Making, 23(1), 248. https://doi.org/10.1186/s12911-023-02350-w
  5. Alzahrani, S., Nadershah, M., Alghamdi, M., Baabdullah, R., Bayoumi, M., Bawajeeh, O., ... & Bayoumi, A. (2025). The use of Apple smartwatches to obtain vital signs readings in surgical patients. Scientific Reports, 15(1), 1-10. https://doi.org/10.1038/s41598-024-84459-0
  6. Navalta, J. W., Montes, J., Bodell, N. G., Salatto, R. W., Manning, J. W., & DeBeliso, M. (2020). Concurrent heart rate validity of wearable technology devices during trail running. Plos one, 15(8), e0238569. https://doi.org/10.1371/journal.pone.0238569
  7. Meana-Llorián, D., García, C. G., G-bustelo, B. C. P., Lovelle, J. M. C., & Garcia-Fernandez, N. (2017). IoFClime: The fuzzy logic and the Internet of Things to control indoor temperature regarding the outdoor ambient conditions. Future Generation Computer Systems, 76, 275-284. https://doi.org/10.1016/j.future.2016.11.020
  8. Smits, E. J., Wenzel, N., & de Bruin, A. (2025). Investigating the Effectiveness of Self-Regulated, Pomodoro, and Flowtime Break-Taking Techniques Among Students. Behavioral Sciences, 15(7), 861. https://doi.org/10.3390/bs15070861
  9. Renacido, J. M. D., Mayordo, E. L., & Biray, E. T. (2025). A Comparative Study Between Pomodoro and Flowtime Techniques Among College Students. International Journal of Multidisciplinary: Applied Business and Education Research, 6(8), 3953-3973. https://doi.org/10.11594/ijmaber.06.08.17
  10. Santiago, C., & Gurat, M. (2023). The Effect of Pomodoro Technique on Student Mendelian Genetics Concept Mastery during Synchronous Remote Learning. International research journal of management, IT and social sciences, 10(4), 233-243. https://doi.org/10.21744/irjmis.v10n4.2287
  11. Litayem, N. (2024, December). Scalable smart home management with ESP32-S3: A low-cost solution for accessible home automation. 2024 International Conference on Computer and Applications (ICCA), 1-7. https://doi.org/10.1109/ICCA62237.2024.10927887
  12. Ouyang, L., Zhu, G., & Zhang, Y. (2025, June). Design of Deepseek Big Model Intelligent Voice Assistant Based on ESP32 Microcontroller. In 2025 IEEE International Conference on Pattern Recognition, Machine Vision and Artificial Intelligence (PRMVAI), 1-5. https://doi.org/10.1109/PRMVAI65741.2025.11108406




DOI: http://dx.doi.org/10.30970/eli.31.3

Refbacks

  • There are currently no refbacks.