EUVIMEDEUVIMED entdecken
Crossref

CxLSTM: Deploying xLSTM-Based Custom Keyword Spotting to Edge Devices

Haitao Cao, Tianyou Xiao, Chao Chang, Luyao Teng, Xi Liu, Dan Deng

Journal of Advanced Computational Intelligence and Intelligent Informatics · 20.11.2026 · 30 (6) · 2084-2101

DOI: 10.20965/jaciii.2026.p2084 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Advanced Computational Intelligence and Intelligent Informatics, online erschienen am 20.11.2026. Verfasst von Haitao Cao und 5 weiteren Autor:innen. Die Arbeit nennt 48 Literaturangaben. Gefördert von: the Guangzhou Municipal Education Bureau, the Collaborative Innovation Center for Intelligent Educational Technology of Guangzhou, the Key Research Platform of Guangzhou Polytechnic University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Keyword spotting (KWS) serves as the critical entry point for voice interaction in intelligent systems. Traditional KWS systems rely on predefined keywords with limited customization, whereas existing custom keyword spotting (CKWS) approaches incur heavy computational overhead, limiting their deployment on resource-limited edge devices. To address these challenges, we present CxLSTM, an optimized extended LSTM module with only 57.6K parameters, 318.6 KB static memory, 593.1 KB peak memory, and 3.54 MACs per inference. Based on CxLSTM, we propose an on-device CKWS method and perform extensive experiments. In offline evaluation, the proposed method obtains an equal error rate (EER) of 0.27 and an area under the ROC curve (AUC) of 0.78 on the public Google Speech Commands V2 dataset. Superior performance is further validated on a private Mandarin dataset, yielding an EER of 0.06 and an AUC of 0.98. Furthermore, in real-world tests for Mandarin CKWS across three scenarios, the proposed method achieves a recognition accuracy above 95% on Android devices, with a false alarm rate below 0.75/24 h and an average latency of 2.1 ms. When deployed on STM32 microcontrollers, it maintains an accuracy of over 92.3%, with moderately increased false alarms (15.8/24 h) and latency (264.7 ms). Despite performance degradation on low-power microcontrollers due to limited hardware resources, the proposed CKWS method still enables efficient voice interaction on edge devices.

Förderung

  • the Guangzhou Municipal Education Bureau
  • the Collaborative Innovation Center for Intelligent Educational Technology of Guangzhou
  • the Key Research Platform of Guangzhou Polytechnic University
  • the Education Department of Guangdong

Zitieren

APA

Cao, H., Xiao, T., Chang, C., Teng, L., Liu, X., & Deng, D. (2026). CxLSTM: Deploying xLSTM-Based Custom Keyword Spotting to Edge Devices. Journal of Advanced Computational Intelligence and Intelligent Informatics, 30(6), 2084-2101. https://doi.org/10.20965/jaciii.2026.p2084

Vancouver

Cao H, Xiao T, Chang C, Teng L, Liu X, Deng D. CxLSTM: Deploying xLSTM-Based Custom Keyword Spotting to Edge Devices. Journal of Advanced Computational Intelligence and Intelligent Informatics. 2026;30(6):2084-2101. doi:10.20965/jaciii.2026.p2084

BibTeX
@article{Cao2026,
  author = {Cao, Haitao and Xiao, Tianyou and Chang, Chao and Teng, Luyao and Liu, Xi and Deng, Dan},
  title = {CxLSTM: Deploying xLSTM-Based Custom Keyword Spotting to Edge Devices},
  journal = {Journal of Advanced Computational Intelligence and Intelligent Informatics},
  year = {2026},
  volume = {30},
  number = {6},
  pages = {2084--2101},
  doi = {10.20965/jaciii.2026.p2084}
}
Kostenlos starten →