Improving Latency and Stability in Edge-Based Voice Assistants Through Memory and Scheduling Optimization
Main Article Content
Abstract
Intelligent voice assistants are now widely used on smartphones and embedded boards, where short response time and stable operation are essential. Heavy computation and limited hardware, however, constrain efficiency. This study tested a dual-path method that applied fine-grained memory control together with asynchronous scheduling. A total of 110 trials were run in both laboratory and office conditions. Results showed that median latency fell by 37.3% and 95th percentile latency by 39.8%. Jitter was reduced by 24.6%, and timeout events dropped by 74% compared with baseline runs. Accuracy remained stable, with word error rate changes not exceeding 0.2 and F1 score changes not exceeding 0.3. The results indicate that combining algorithm-level and system-level methods gives stronger benefits than using them alone. The study also reports jitter and timeout metrics, which are often not considered in related work. These findings suggest that dual-path optimization can support efficient and reliable deployment of voice assistants on edge devices. The main limits are the small number of device types, short test periods, and the use of only English speech. Future work should extend to multilingual datasets, longer trials, and secure execution tests.
Article Details
Section
How to Cite
References
1. R. Wolniak, and W. Grebski, "The usage of smart voice assistant in smart home," Zeszyty Naukowe. Organizacja i Zarzadzanie/Politechnika Slaska; Silesian University of Technology Publishing House, pp. 701-710, 2023.
2. M. Yuan, W. Qin, J. Huang, and Z. Han, "A robotic digital construction workflow for puzzle-assembled freeform architectural components using castable sustainable materials," 2025. doi: 10.2139/ssrn.5433279
3. F. Chen, L. Yue, P. Xu, H. Liang, and S. Li, "Research on the efficiency improvement algorithm of electric vehicle energy recovery system based on GaN power module," 2025. doi: 10.20944/preprints202506.1323.v1
4. C. Wu, J. Zhu, and Y. Yao, "Identifying and optimizing performance bottlenecks of logging systems for augmented reality platforms," 2025. doi: 10.20944/preprints202509.0357.v1
5. Z. Li, "Traffic density road gradient and grid composition effects on electric vehicle energy consumption and emissions," Innovations in Applied Engineering and Technology, pp. 1-8, 2023.
6. K. L. Chiu, "Transparent deployment of machine learning models on many-accelerator architectures (Doctoral dissertation, Columbia University)," 2025.
7. K. Xu, Q. Wu, Y. Lu, Y. Zheng, W. Li, X. Tang, and X. Sun, "Meatrd: Multimodal anomalous tissue region detection enhanced with spatial transcriptomics," In Proceedings of the AAAI Conference on Artificial Intelligence, April, 2025, pp. 12918-12926. doi: 10.1609/aaai.v39i12.33409
8. P. Mattson, C. Cheng, G. Diamos, C. Coleman, P. Micikevicius, D. Patterson, and M. Zaharia, "MLPerf training benchmark," Proceedings of Machine Learning and Systems, vol. 2, pp. 336-349, 2020.
9. Y. Yang, X. Xie, X. Wang, H. Zhang, C. Yu, X. Xiong, and F. Baik, "Impact of target and tool visualization on depth perception and usability in optical see-through AR," 2025. doi: 10.1109/ismar-adjunct68609.2025.00148
10. C. Zhang, H. Yu, X. Luo, W. Yin, J. Huang, X. Liu, and Z. Liu, “CitySense RAG: Personalized urban mobility recommendations via streetscape perception and multi-source semantics,” in press, 2025.
11. V. M. Souza, D. M. dos Reis, A. G. Maletzke, and G. E. Batista, "Challenges in benchmarking stream learning algorithms with real-world data," Data Mining and Knowledge Discovery, vol. 34, no. 6, pp. 1805-1858, 2020.
12. Y. Guo, and S. Yang, "Noise effects on purity and quantum entanglement in terms of physical implementability," npj Quantum Information, vol. 9, no. 1, p. 11, 2023. doi: 10.21203/rs.3.rs-1878672/v1
13. W. Sun, “Integration of market-oriented development models and marketing strategies in real estate,” European Journal of Business, Economics & Management, vol. 1, no. 3, pp. 45–52, 2025.
14. H. Chen, X. Ma, Y. Mao, and P. Ning, "Research on low latency algorithm optimization and system stability enhancement for intelligent voice assistant," 2025. doi: 10.1109/icecai66283.2025.11170668
15. M. Sabahi, A. Safari, and M. Nazari-Heris, "Design and implementation of a cost-effective practical single-phase power quality analyzer using pyboard microcontroller and python-to-python interface," The Journal of Engineering, vol. 2024, no. 2, p. e12360, 2024.
16. Q. Li, J. Zheng, A. Tsai, and Q. Zhou, "Robust endpoint detection and energy normalization for real-time speech and speaker recognition," IEEE Transactions on Speech and Audio Processing, vol. 10, no. 3, pp. 146-157, 2002.
17. Y. Huang, W. He, Y. Kantaros, and S. Zeng, "Spatiotemporal co-design enabling prioritized multi-agent motion planning," In 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), October, 2024, pp. 10281-10288. doi: 10.1109/iros58592.2024.10801559
18. S. Ghose, H. Lee, and J. F. Martínez, "Improving memory scheduling via processor-side load criticality information," In Proceedings of the 40th Annual International Symposium on Computer Architecture, June, 2013, pp. 84-95. doi: 10.1145/2508148.2485930
19. L. T. Clark, V. De, I. Verbauwhede, R. David, S. Pillement, O. Sentieys, and A. Macii, "Low-power processors and memories," In Low-Power Processors and Systems on Chips., 2006.