Ultra-low power on-chip learning of speech commands with phase-change memories
Embedding artificial intelligence at the edge (edge-AI) is an elegant solution to tackle the power and latency issues in the rapidly expanding Internet of Things. As edge devices typically spend most of their time in sleep mode and only wake-up infrequently to collect and process sensor data, non-volatile in-memory computing (NVIMC) is a promising approach to design the next generation of edge-AI devices. Recently, we proposed an NVIMC-based neuromorphic accelerator using the phase change memories (PCMs), which we call as Raven. In this work, we demonstrate the ultra-low-power on-chip training and inference of speech commands using Raven. We showed that Raven can be trained on-chip with power consumption as low as 30~uW, which is suitable for edge applications. Furthermore, we showed that at iso-accuracies, Raven needs 70.36x and 269.23x less number of computations to be performed than a deep neural network (DNN) during inference and training, respectively. Owing to such low power and computational requirements, Raven provides a promising pathway towards ultra-low-power training and inference at the edge.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Deep Convolutional Spiking Neural Networks for Keyword Spotting
This paper investigates the use of deep convolutional spiking neural networks (SNN) for keyword spotting (KWS) and wakeword detection tasks. The brain-inspired SNN mimic the spike-based information processing of biologic…
Keyword SpottingA Bin Encoding Training of a Spiking Neural Network-based Voice Activity Detection
Advances of deep learning for Artificial Neural Networks(ANNs) have led to significant improvements in the performance of digital signal processing systems implemented on digital chips. Although recent progress in low-po…
Action DetectionActivity DetectionA CMOS SoC for Wireless Ultrasonic Power/Data Transfer and SHM Measurements on Structures
This paper describes a highly-integrated CMOS system-on-chip (SoC) for active structural health monitoring (SHM). The chip integrates ultrasonic power and bidirectional half-duplex data transfer, a power management unit …
ManagementStructural Health MonitoringDecentralized LLM-Driven Coordination of Acoustic Robots for Contactless Object Manipulation
Natural language interfaces can simplify interaction with multi-robot systems, especially when non-expert users need to issue high-level commands. Acoustic manipulation using ultrasonic phased arrays also enables contact…
Speech RecognitionSemantic ParsingLLMET: Enabling Cross-Layer Evaluation of Emerging M3D Memories for Energy-Efficient LLM Serving
The energy consumption of Large Language Model (LLM) serving is becoming a major system challenge as deployment scales, driven by hardware power and thermal constraints and rising electricity costs. A key contributor to …