在嵌入式芯片上实现低功耗实时语音关键词识别。
Implementing Keyword Spotting on the MCUX947 Microcontroller with Integrated NPU
- 用MFCC+CNN结合量化训练压缩模型
- 推理速度提升59倍，准确率达97.06%
- 适合物联网设备的轻量级语音交互
本文提出一种在NXP MCXN947微控制器（集成神经处理单元NPU）上实现的关键词识别（KWS）系统，支持资源受限设备上的实时语音交互。系统采用梅尔频率倒谱系数（MFCC）特征提取与卷积神经网络（CNN）分类器，通过量化感知训练优化模型，在保持高精度的同时显著减小模型体积。实验表明，利用NPU相比纯CPU执行可实现59倍的推理速度提升，模型大小仅为30.58 KB，准确率达到97.06%，验证了在嵌入式平台实现高效、低功耗语音接口的可行性。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。