提出抗歌声干扰的语音活动检测模型，提升复杂场景下语音识别准确性
Robust Speech Activity Detection in the Presence of Singing Voice
- 通过控制语音与歌声样本比例训练模型，增强两者区分能力
- 在多音乐风格数据集上实现0.919的AUC，有效拒绝不属于语音的歌声
- 适合语音识别、对话增强等需精准识别语音的场景
语音活动检测（SAD）系统常将歌声误判为语音，影响对话增强和自动语音识别性能。本文提出歌唱鲁棒语音活动检测（SR-SAD），一种神经网络模型，可在存在歌声时仍准确检测语音。主要贡献包括：(i) 使用可控比例的语音与歌声样本进行训练，提升二者的区分能力；(ii) 设计计算高效的模型，在保持鲁棒性的同时降低推理耗时；(iii) 提出新评估指标，专门用于衡量混合语音-歌声场景下的SAD鲁棒性。在涵盖多种音乐风格的挑战性数据集上的实验表明，SR-SAD在保持高语音检测精度（AUC = 0.919）的同时能有效排除歌声。通过显式学习区分语音与歌声，该模型显著提升了混合场景中的语音活动检测可靠性。
原文摘要 · Abstract (English)
Speech Activity Detection (SAD) systems often misclassify singing as speech, leading to degraded performance in applications such as dialogue enhancement and automatic speech recognition. We introduce Singing-Robust Speech Activity Detection ( SR-SAD ), a neural network designed to robustly detect speech in the presence of singing. Our key contributions are: i) a training strategy using controlled ratios of speech and singing samples to improve discrimination, ii) a computationally efficient model that maintains robust performance while reducing inference runtime, and iii) a new evaluation metric tailored to assess SAD robustness in mixed speech-singing scenarios. Experiments on a challenging dataset spanning multiple musical genres show that SR-SAD maintains high speech detection accuracy (AUC = 0.919) while rejecting singing. By explicitly learning to distinguish between speech and singing, SR-SAD enables more reliable SAD in mixed speech-singing scenarios.
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。