提出一种低延迟的关键词检测新结构，适合实时语音处理。
Cumsum-Composable Phase Transport for Low-Cost Streaming Keyword Spotting
- 用复数通道与酉旋转实现流式音频的高效时序传输
- 24.8K参数模型达96.8%准确率，训练速度比传统方法快1.07倍
- 适合资源受限设备部署，特别适合低延迟关键词识别场景
状态空间序列模型因保持紧凑的循环状态而适用于流式语音，但扫描式训练核在短音频任务中常有不利常数。本文研究了累积和可组合相位传输，一种专为关键词检测设计的流式原生时序层。每层将声学帧投影到复数通道，通过学习的酉旋转进行传输，使用前缀差分累积有限窗口，并应用门控残差更新。相同的前缀表示支持普通累积和的精确批处理训练，以及每帧仅一次前缀更新的精确在线推理。酉传输是关键约束：逆旋转范数为一，使前缀项保持良好条件性，同时通过窗口或块读出提供记忆。在包含12类标签的Google Speech Commands v2数据集上，mel+cumsum模型在紧凑基线中保持竞争力。最强单种子运行达到97.3%测试准确率；51.6K参数的共享模型也达97.3%，24.8K共享模型达96.8%，对比25.6K MelCNNMaxPool基线的97.1%。在匹配的cumsum与scan基准测试中，cumsum+window在94.82%准确率下，相比94.33%的同时，训练速度快1.07倍，单样本延迟从7.09毫秒降至5.01毫秒（Tesla T4）。这些结果表明，cumsum相位传输是一种简洁、低成本的流式关键词检测时序基础模块。
原文摘要 · Abstract (English)
State-space sequence models are attractive for streaming speech because they maintain compact recurrent state, but scan-style training kernels can have unfavorable constants for short audio tasks. We study cumsum-composable phase transport, a streaming-native temporal layer for keyword spotting. Each layer projects acoustic frames to complex channels, transports them by learned unitary rotations, accumulates a finite window using prefix differences, and applies a gated residual update. The same prefix representation gives exact batched training with ordinary cumulative sums and exact online inference with one prefix update per frame. Unitary transport is the key constraint: inverse rotations have norm one, keeping prefix terms well conditioned while memory is supplied by windows or block readouts. On Google Speech Commands v2 with 12 labels, mel+cumsum models retain competitive accuracy with compact baselines. The strongest single-seed run reaches 97.3\% test accuracy; a 51.6K-parameter tied model also reaches 97.3\%, and a 24.8K tied model reaches 96.8\% versus 97.1\% for a 25.6K MelCNNMaxPool baseline. In a matched cumsum-versus-scan benchmark, cumsum+window gives comparable accuracy, 94.82\% versus 94.33\%, while training 1.07x faster and reducing single-example latency from 7.09 ms to 5.01 ms on a Tesla T4. These results support cumsum phase transport as a simple low-cost temporal primitive for streaming keyword spotting.
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。