mlsanityData to 14 Oct 2026

A Fast Network Exploration Strategy to Profile Low Energy Consumption for Keyword Spotting

arXiv 2202.02361 ↗First posted 4 Feb 2022cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

Keyword Spotting nowadays is an integral part of speech-oriented user interaction targeted for smart devices. To this extent, neural networks are extensively used for their flexibility and high accuracy. However, coming up with a suitable configuration for both accuracy requirements and hardware deployment is a challenge. We propose a regression-based network exploration technique that considers the scaling of the network filters (s) and quantization (q) of the network layers, leading to a friendly and energy-efficient configuration for FPGA hardware implementation. We experiment with different combinations of NNscriptstylelangle q,,srangle displaystyle on the FPGA to profile the energy consumption of the deployed network so that the user can choose the most energy-efficient network configuration promptly. Our accelerator design is deployed on the Xilinx AC 701 platform and has at least 2.1× and 4× improvements on energy and energy efficiency results, respectively, compared to recent hardware implementations for keyword spotting.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 8 of its 20 references that are in the map. First posted on arXiv 4 Feb 2022.

Builds on

8 of 20 references.

References · 8

  1. 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 752
  2. 2017Hello Edge: Keyword Spotting on Microcontrollerscited by 110
  3. 2020Streaming keyword spotting on mobile devicescited by 47
  4. 2016Ternary Neural Networks for Resource-Efficient AI Applicationscited by 33
  5. 2017Convolutional Recurrent Neural Networks for Small-Footprint Keyword Spottingcited by 32
  6. 2019Small-Footprint Keyword Spotting on Raw Audio Data with Sinc-Convolutionscited by 20
  7. 2020Neural Architecture Search For Keyword Spottingcited by 11
  8. 2021Tiny-CRNN: Streaming Wakeword Detection In A Low Footprint Settingcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Hardware-Software Co-Design of Scalable, Energy-Efficient Analog Recurrent Computationscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022Sub-mW Keyword Spotting on an MCU: Analog Binary Feature Extraction and Binary Neural Networks
  2. 2022A 14uJ/Decision Keyword Spotting Accelerator with In-SRAM-Computing and On Chip Learning for Customization
  3. 2026End-to-End Keyword Spotting on FPGA Using Graph Neural Networks with a Neuromorphic Auditory Sensor
  4. 2022Sub 8-Bit Quantization of Streaming Keyword Spotting Models for Embedded Chipsets
  5. 2026EdgeSpot: Efficient and High-Performance Few-Shot Model for Keyword Spotting
  6. 2022Keyword Spotting System and Evaluation of Pruning and Quantization Methods on Low-power Edge Microcontrollers
  7. 2025Implementing Keyword Spotting on the MCUX947 Microcontroller with Integrated NPU
  8. 2024Micro-power spoken keyword spotting on Xylo Audio 2