Effects of interpulse-interval variation on deep-learning classification of bat vocalizations
作者: Welmoed R. Eversteijn, Burooj Ghani, A. Leonie Baier, Dan Stowell
分类: cs.LG, eess.AS
发布日期: 2026-10-01
备注: 10 pages, 3 figures
💡 一句话要点
研究间隔脉冲变化对蝙蝠声学分类的影响
🎯 匹配领域: 支柱八:物理动画 (Physics-based Animation)
关键词: 蝙蝠声学 深度学习 物种分类 时间上下文 变换器模型 卷积神经网络 生态监测
📋 核心要点
- 现有方法对蝙蝠声学分类中时间上下文的利用不足,特别是间隔脉冲间隔(IPI)的变化对物种区分的贡献尚不明确。
- 本文通过构建自然IPI和标准化IPI条件的数据集,比较了变换器模型与卷积神经网络在蝙蝠声学分类中的表现。
- 实验结果显示,PaSST在两种条件下的准确率均优于EfficientNet,且自然IPI条件下的训练模型在测试集上表现更佳。
📝 摘要(中文)
本文探讨了时间上下文在自动化蝙蝠物种分类中的作用,特别是间隔脉冲间隔(IPI)的变化是否提供物种区分信息,以及变换器模型是否比卷积神经网络更敏感。研究使用了来自欧洲蝙蝠录音的两个匹配数据集,分别为自然IPI条件和标准化IPI条件。通过对EfficientNet-B0和PaSST的微调与评估,发现自然IPI变化对分类的贡献有限,且变换器模型与卷积网络在使用自然IPI时效果相近。尽管如此,标准化条件下的结果未能完全转移到自然录音上。
🔬 方法详解
问题定义:本文旨在探讨间隔脉冲间隔(IPI)变化对蝙蝠物种分类的影响,现有方法未能充分利用时间上下文信息,导致分类性能受限。
核心思路:通过比较自然IPI与标准化IPI条件下的模型表现,研究不同模型对时间上下文的敏感性,特别是变换器模型与卷积神经网络的差异。
技术框架:研究使用了两个数据集:自然IPI条件保留原始呼叫时序,标准化IPI条件则将呼叫间隔设定为50毫秒。模型包括EfficientNet-B0和PaSST,分别在这两种条件下进行微调与评估。
关键创新:本研究的创新在于系统性地评估了IPI变化对蝙蝠分类的影响,并比较了不同模型在处理时间上下文信息时的表现差异。
关键设计:在实验中,EfficientNet和PaSST的训练与评估采用了不同的IPI条件,损失函数和网络结构均基于各自的设计理念进行优化。
🖼️ 关键图片
📊 实验亮点
实验结果表明,PaSST在自然IPI条件下的准确率为71%,而在标准化IPI条件下为70%;EfficientNet的准确率从47%提升至57%。在跨条件评估中,使用自然IPI训练的模型在自然IPI测试集上的表现优于标准化训练模型,显示出标准化条件下的结果不完全适用于自然录音。
🎯 应用场景
该研究为蝙蝠物种分类提供了新的视角,尤其是在生态监测和生物多样性保护领域。通过改进分类模型的设计,可以更有效地识别和监测蝙蝠种群,为生态学研究和保护策略提供数据支持。
📄 摘要(原文)
Temporal context may aid automated bat-species classification, but the contribution of specific features remains unclear. We investigated whether variation in the interpulse interval (IPI)-the time between consecutive call onsets-provides species-discriminative information and whether transformer-based models are more sensitive to this information than convolutional neural networks. We created two matched datasets from European bat recordings: a natural-IPI condition retaining the original call timing and a normalized-IPI condition in which call onsets were spaced at 50-ms intervals. EfficientNet-B0 and PaSST were fine-tuned and evaluated within each condition. In an additional experiment, each architecture was trained separately on natural-IPI and normalized-IPI recordings, and evaluated on the same natural-IPI test set. Finally, the pretrained classifiers BatDetect2 and BAT were evaluated on both conditions. Within-condition IPI normalization had model-dependent effects. PaSST accuracy differed little between the natural-IPI ($71 \pm 2.3\%$) and normalized-IPI ($70 \pm 6.3\%$) conditions, whereas EfficientNet accuracy increased from $47 \pm 4.7\%$ to $57 \pm 3.9\%$. PaSST exceeded EfficientNet under both conditions. In the cross-condition evaluation, models trained on natural-IPI recordings outperformed those trained on normalized-IPI recordings on the natural-IPI test set: accuracy decreased from 54% to 50% for EfficientNet and from 65% to 57% for PaSST. BatDetect2 and BAT differed little between IPI conditions. Overall, we found limited support for the hypotheses that natural IPI variation contributes substantially to bat-species classification and that it is used more effectively by transformer-based than CNN-based models. Nevertheless, the cross-condition performance decrease shows that results obtained under normalized conditions may not transfer fully to natural recordings.