A nine-transistor current-mode bistable memory cell in 180 nm CMOS is presented with independent tuning of threshold, hysteresis, and gain, shown via schematic simulations for spike-based logic gates and recurrent neural units.
Hello edge: Keyword spotting on microcontrollers
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
abstract
Keyword spotting (KWS) is a critical component for enabling speech based user interactions on smart devices. It requires real-time response and high accuracy for good user experience. Recently, neural networks have become an attractive choice for KWS architecture because of their superior accuracy compared to traditional speech processing algorithms. Due to its always-on nature, KWS application has highly constrained power budget and typically runs on tiny microcontrollers with limited memory and compute capability. The design of neural network architecture for KWS must consider these constraints. In this work, we perform neural network architecture evaluation and exploration for running KWS on resource-constrained microcontrollers. We train various neural network architectures for keyword spotting published in literature to compare their accuracy and memory/compute requirements. We show that it is possible to optimize these neural network architectures to fit within the memory and compute constraints of microcontrollers without sacrificing accuracy. We further explore the depthwise separable convolutional neural network (DS-CNN) and compare it against other neural network architectures. DS-CNN achieves an accuracy of 95.4%, which is ~10% higher than the DNN model with similar number of parameters.
citation-role summary
baseline 1
citation-polarity summary
roles
baseline 1polarities
baseline 1representative citing papers
NoNN partitions a teacher model into disjoint compressed students via network science for distributed IoT inference, matching teacher accuracy with far lower per-device memory and communication.
Non-IID data causes up to 55% accuracy loss in federated learning due to weight divergence measured by earth mover's distance; 5% globally shared data recovers 30% accuracy on CIFAR-10.
EdgeSpike delivers 91.4% mean accuracy on five sensing tasks with 31x lower energy on neuromorphic hardware and 6.3x longer battery life in a seven-month field deployment compared to conventional CNNs.
Dendritic models using Perforated Backpropagation reach 0.933 test accuracy with 1500 parameters on keyword spotting, beating a baseline of 0.921 accuracy that needs roughly 4000 parameters.
citing papers explorer
-
A Fully Tunable Ultra-Low Power Current-Mode Memory Cell in Standard CMOS Technology
A nine-transistor current-mode bistable memory cell in 180 nm CMOS is presented with independent tuning of threshold, hysteresis, and gain, shown via schematic simulations for spike-based logic gates and recurrent neural units.
-
Memory- and Communication-Aware Model Compression for Distributed Deep Learning Inference on IoT
NoNN partitions a teacher model into disjoint compressed students via network science for distributed IoT inference, matching teacher accuracy with far lower per-device memory and communication.
-
Federated Learning with Non-IID Data
Non-IID data causes up to 55% accuracy loss in federated learning due to weight divergence measured by earth mover's distance; 5% globally shared data recovers 30% accuracy on CIFAR-10.
-
EdgeSpike: Spiking Neural Networks for Low-Power Autonomous Sensing in Edge IoT Architectures
EdgeSpike delivers 91.4% mean accuracy on five sensing tasks with 31x lower energy on neuromorphic hardware and 6.3x longer battery life in a seven-month field deployment compared to conventional CNNs.
-
Perforated Neural Networks for Keyword Spotting
Dendritic models using Perforated Backpropagation reach 0.933 test accuracy with 1500 parameters on keyword spotting, beating a baseline of 0.921 accuracy that needs roughly 4000 parameters.
- Hardware-Software Co-Design of Scalable, Energy-Efficient Analog Recurrent Computations