通过子序列匹配提升短关键词在边缘设备上的识别效率
SLiCK: Exploiting Subsequences for Length-Constrained Keyword Spotting
- 将关键词识别转为长度约束问题，避免变长文本聚合
- 在Libriphrase硬集上AUC达94.9，误报率降低至11.1%
- 适合资源受限场景下的实时关键词检测应用
用户自定义关键词在资源受限的边缘设备上进行检测具有挑战性。然而，关键词通常存在最大长度限制，这一特性在以往研究中未被充分利用。我们对关键词长度分布的分析表明，用户定义的关键词检测可视为长度约束问题，从而无需对变长文本进行聚合处理。基于此，我们提出高效关键词检测方法SLiCK（Exploiting Subsequences for Length-Constrained Keyword Spotting）。引入子序列级匹配机制，以更细粒度学习音频-文本关联，从而更有效地区分发音相似的关键词。SLiCK采用多任务学习框架，包含两个模块：匹配器（执行话语级匹配与新型子序列级匹配任务）和编码器（执行音素识别任务）。在Libriphrase hard数据集上，该方法将基线AUC从88.52提升至94.9，等错误率（EER）从18.82降至11.1。
原文摘要 · Abstract (English)
User-defined keyword spotting on a resource-constrained edge device is challenging. However, keywords are often bounded by a maximum keyword length, which has been largely under-leveraged in prior works. Our analysis of keyword-length distribution shows that user-defined keyword spotting can be treated as a length-constrained problem, eliminating the need for aggregation over variable text length. This leads to our proposed method for efficient keyword spotting, SLiCK (exploiting Subsequences for Length-Constrained Keyword spotting). We further introduce a subsequence-level matching scheme to learn audio-text relations at a finer granularity, thus distinguishing similar-sounding keywords more effectively through enhanced context. In SLiCK, the model is trained with a multi-task learning approach using two modules: Matcher (utterance-level matching task, novel subsequence-level matching task) and Encoder (phoneme recognition task). The proposed method improves the baseline results on Libriphrase hard dataset, increasing AUC from $88.52$ to $94.9$ and reducing EER from $18.82$ to $11.1$.
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。