用大模型生成易混淆关键词语音，提升自定义语音唤醒准确率
LLM-Synth4KWS: Scalable Automatic Generation and Synthesis of Confusable Data for Custom Keyword Spotting
- 用大语言模型生成易混淆关键词对，结合多种语音合成技术扩增数据
- 在语音命令数据集上使整体识别率提升3.7%，易混淆词识别率提升11.3%
- 零人工成本、高可扩展，适合需要精准唤醒的智能设备研发
自定义关键词唤醒（KWS）通过比对语音注册与输入音频的嵌入向量实现。当前先进模型通常基于训练集中随机采样的关键词进行对比学习，难以处理易混淆词（如"blue"与"glue"）。本文提出一种新方法：利用大语言模型（LLM）生成并分组易混淆关键词，再通过文本到语音（TTS）引擎合成多样风格的语音信号用于训练。为更真实评估用户体验，引入新指标c-AUC（易混淆组DET曲线平均面积）。该方法具有高可扩展性与零人工成本优势，在Speech Commands测试集上使AUC提升3.7%，c-AUC提升11.3%。
原文摘要 · Abstract (English)
Custom keyword spotting (KWS) allows detecting user-defined spoken keywords from streaming audio. This is achieved by comparing the embeddings from voice enrollments and input audio. State-of-the-art custom KWS models are typically trained contrastively using utterances whose keywords are randomly sampled from training dataset. These KWS models often struggle with confusing keywords, such as "blue" versus "glue". This paper introduces an effective way to augment the training with confusable utterances where keywords are generated and grouped from large language models (LLMs), and speech signals are synthesized with diverse speaking styles from text-to-speech (TTS) engines. To better measure user experience on confusable KWS, we define a new northstar metric using the average area under DET curve from confusable groups (c-AUC). Featuring high scalability and zero labor cost, the proposed method improves AUC by 3.7% and c-AUC by 11.3% on the Speech Commands testing set.
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。