EdgeSpot在边缘设备上实现高效高精度关键词识别，10次射击下准确率提升8.3个百分点。
EdgeSpot: Efficient and High-Performance Few-Shot Model for Keyword Spotting
- 结合优化的声学骨干与轻量自注意力机制，提升小样本学习效率
- 10次射击时在1%误报率下准确率达82.0%，比基线提升8.3个百分点
- 模型仅需29.4M MACs和128k参数，适合资源受限边缘设备
我们提出一种面向边缘设备的高效少样本关键词识别模型EdgeSpot，其采用优化的基于BC-ResNet的声学主干网络，搭配可训练的逐通道能量归一化前端和轻量级时序自注意力模块。训练过程中引入自监督教师模型，并使用子中心ArcFace损失进行知识蒸馏。实验表明，EdgeSpot在固定误报率（FAR）下持续优于强基线模型。最大版本EdgeSpot-4在10次射击、1% FAR条件下，准确率从73.7%提升至82.0%，仅需29.4M MACs与128k参数。
原文摘要 · Abstract (English)
We introduce an efficient few-shot keyword spotting model for edge devices, EdgeSpot, that pairs an optimized version of a BC-ResNet-based acoustic backbone with a trainable Per-Channel Energy Normalization frontend and lightweight temporal self-attention. Knowledge distillation is utilized during training by employing a self-supervised teacher model, optimized with Sub-center ArcFace loss. This study demonstrates that the EdgeSpot model consistently provides better accuracy at a fixed false-alarm rate (FAR) than strong BC-ResNet baselines. The largest variant, EdgeSpot-4, improves the 10-shot accuracy at 1% FAR from 73.7% to 82.0%, which requires only 29.4M MACs with 128k parameters.
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。