EUVIMEDEUVIMED entdecken
Crossref

A 0.05-mm 2 2.91-nJ/Decision Keyword-Spotting (KWS) Chip Featuring an Always-Retention 5T-SRAM in 28-nm CMOS

Fei Tan, Wei-Han Yu, Ka-Fai Un, Rui P. Martins, Pui-In Mak

IEEE Journal of Solid-State Circuits · 2024-02 · 59 (2) · 626-635

DOI: 10.1109/jssc.2023.3291376 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Journal of Solid-State Circuits, 2024-02. Verfasst von Fei Tan und 4 weiteren Autor:innen. Die Arbeit wurde bisher 29-mal zitiert (Crossref) und nennt 26 Literaturangaben. Gefördert von: Macao Science and Technology Development Fund (FDCT), Macau, Successive-Approximation-Register, University of Macau.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This article reports a keyword-spotting (KWS) chip for voice-controlled devices. It features a number of techniques to enhance the performance, area, and power efficiencies: 1) a fast-sampling convolutional neural network (FS-CNN) that eliminates the power-hungry feature extractors and reduces the decision latency; 2) an always-retention 5T-SRAM that features word-voltage switches to reduce the leakage power and single bitline (BL) operation to halve the SRAM read power compared to the typical 6T-SRAM; and 3) a high-resolution sparsity-aware computing (HR-SAC) unit that enhances the precision and output swing of the multiply–accumulate (MAC) computation. Benchmarking with the state-of-the-art, our KWS chip prototyped in 28-nm CMOS scores a $>$ 90% accuracy for the 11-class Google speech command dataset (GSCD) at 2.91 $\mu \text{W}$ , which corresponds to a 2.91-nJ energy/decision. The achieved latency is 2 ms/decision, and the core area is 0.05 ${\mathrm{ mm}}^{2}$ , including the full KWS model.

Förderung

  • Macao Science and Technology Development Fund (FDCT), Macau, Successive-Approximation-Register
  • University of Macau

Zitieren

APA

Tan, F., Yu, W., Un, K., Martins, R. P., & Mak, P. (2024). A 0.05-mm 2 2.91-nJ/Decision Keyword-Spotting (KWS) Chip Featuring an Always-Retention 5T-SRAM in 28-nm CMOS. IEEE Journal of Solid-State Circuits, 59(2), 626-635. https://doi.org/10.1109/jssc.2023.3291376

Vancouver

Tan F, Yu W, Un K, Martins RP, Mak P. A 0.05-mm 2 2.91-nJ/Decision Keyword-Spotting (KWS) Chip Featuring an Always-Retention 5T-SRAM in 28-nm CMOS. IEEE Journal of Solid-State Circuits. 2024;59(2):626-635. doi:10.1109/jssc.2023.3291376

BibTeX
@article{Tan2024,
  author = {Tan, Fei and Yu, Wei-Han and Un, Ka-Fai and Martins, Rui P. and Mak, Pui-In},
  title = {A 0.05-mm 2 2.91-nJ/Decision Keyword-Spotting (KWS) Chip Featuring an Always-Retention 5T-SRAM in 28-nm CMOS},
  journal = {IEEE Journal of Solid-State Circuits},
  year = {2024},
  volume = {59},
  number = {2},
  pages = {626--635},
  doi = {10.1109/jssc.2023.3291376}
}
Kostenlos starten →