AI & ML
Implementing Keyword Spotting on the MCXN947 Microcontroller with Integrated NPU
arXiv cs.HC0 views
arXiv:2506.08911v2 Announce Type: replace Abstract: This paper presents a keyword spotting (KWS) system implemented on the NXP MCXN947 microcontroller with an integrated Neural Processing Unit (NPU), enabling real-time voice interaction on resource-constrained devices. The system combines MFCC feature extraction with a CNN classifier, optimized using Quantization Aware Training to reduce model size with minimal accuracy drop. Experimental results demonstrate a 59x speedup in inference time when leveraging the NPU compared to CPU-only execution, achieving 97.06% accuracy with a model size of 30.58 KB, demonstrating the feasibility of efficient, low-power voice interfaces on embedded platforms.
Source：https://arxiv.org/abs/2506.08911PreviousA State-of-the-Art Review of Computational Models for Analyzing Longitudinal Wearable Sensor Data in HealthcareNextMulti-Agent Home Energy Management Assistant
Related posts
Geometric Functional Inequalities for Markov Processes: An Invitation via the Langevin Diffusion
0arXiv stat.CO
A Unified Coordinate Descent Framework for Penalized Longitudinal Log-Contrast Quantile Regression
0arXiv stat.CO
Fast Functional Clustering via Closed-Form Covariance Updates
0arXiv stat.CO
marp: Model-Averaged Inference for Parametric Renewal Processes in R
0arXiv stat.CO
dsp: An R Package for Bayesian Time Series Modeling with the Dynamic Shrinkage Process
0arXiv stat.CO
Machine Learning Fundamentals: Understanding How Machines Learn from Data
0dev.to Discuss