mlsanityData to 7 Oct 2026

Sub-mW Keyword Spotting on an MCU: Analog Binary Feature Extraction and Binary Neural Networks

arXiv 2201.03386 ↗First posted 10 Jan 2022cs.SD cs.AI cs.ARcited by 5

Not on the map. Why?

Abstract (arXiv)

Keyword spotting (KWS) is a crucial function enabling the interaction with the many ubiquitous smart devices in our surroundings, either activating them through wake-word or directly as a human-computer interface. For many applications, KWS is the entry point for our interactions with the device and, thus, an always-on workload. Many smart devices are mobile and their battery lifetime is heavily impacted by continuously running services. KWS and similar always-on services are thus the focus when optimizing the overall power consumption. This work addresses KWS energy-efficiency on low-cost microcontroller units (MCUs). We combine analog binary feature extraction with binary neural networks. By replacing the digital preprocessing with the proposed analog front-end, we show that the energy required for data acquisition and preprocessing can be reduced by 29x, cutting its share from a dominating 85% to a mere 16% of the overall energy consumption for our reference KWS application. Experimental evaluations on the Speech Commands Dataset show that the proposed system outperforms state-of-the-art accuracy and energy efficiency, respectively, by 1% and 4.3x on a 10-class dataset while providing a compelling accuracy-energy trade-off including a 2% accuracy drop for a 71x energy reduction.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 23 of its 52 references that are in the map. First posted on arXiv 10 Jan 2022.

Builds on

23 of 52 references.

References · 23

  1. 2017Attention Is All You Needcited by 59,932
  2. 2016XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networkscited by 829
  3. 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 741
  4. 2016CNN Architectures for Large-Scale Audio Classificationcited by 701
  5. 2016DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradientscited by 557
  6. 2017Incremental Network Quantization: Towards Lossless CNNs with Low-Precision Weightscited by 211
  7. 2015Fixed Point Quantization of Deep Convolutional Networkscited by 141
  8. 2017Hello Edge: Keyword Spotting on Microcontrollerscited by 108
  9. 2016Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1cited by 102
  10. 2018A multi-device dataset for urban acoustic scene classificationcited by 75
  11. 2017Extremely Low Bit Neural Network: Squeeze the Last Bit Out with ADMMcited by 60
  12. 2017Deep Learning for IoT Big Data and Streaming Analytics: A Surveycited by 59
11 more
  1. 2017A Cascade Architecture for Keyword Spotting on Mobile Devicescited by 15
  2. 2021Sound Event Detection with Binary Neural Networks on Tightly Power-Constrained IoT Devicescited by 10
  3. 2020CUTIE: Beyond PetaOp/s/W Ternary DNN Inference Acceleration with Better-than-Binary Energy Efficiencycited by 8
  4. 2020Compact recurrent neural networks for acoustic event detection on low-energy low-complexity platformscited by 6
  5. 2015Online Keyword Spotting with a Character-Level Recurrent Neural Networkcited by 5
  6. 2017Efficient Convolutional Neural Network For Audio Event Detectioncited by 5
  7. 2019Additive Noise Annealing and Approximation Properties of Quantized Neural Networkscited by 5
  8. 2016An End-to-End Architecture for Keyword Spotting and Voice Activity Detectioncited by 4
  9. 2019An Optimized Recurrent Unit for Ultra-Low-Power Keyword Spottingcited by 4
  10. 2018Binary Weighted Memristive Analog Deep Neural Network for Near-Sensor Edge Processingcited by 2
  11. 2017Mist Computing: Principles, Trends and Future Directioncited by 1

Built on by

Cited by year

Citing papers per year: 2024: 1; 2025: 4.

5 papers.

  1. 2024xTern: Energy-Efficient Ternary Neural Network Inference on RISC-V-Based Edge Systemscited by 4
  2. 2025Efficient Continual Learning in Keyword Spotting using Binary Neural Networkscited by 2
  3. 2025Adaptive Noise Resilient Keyword Spotting Using One-Shot Learningcited by 1
  4. 2025Implementing Keyword Spotting on the MCUX947 Microcontroller with Integrated NPUcited by 1
  5. 2025LearnAFE: Circuit-Algorithm Co-design Framework for Learnable Audio Analog Front-Endcited by 0