arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2110.09600cs.SDeess.AS
收藏

谁说了算?重新思考音频的少样本学习

Who calls the shots? Rethinking Few-Shot Learning for Audio

New York University(纽约大学);Adobe Research(奥多比研究院);New Jersey Institute of Technology(新泽西理工学院)

Yu Wang, Nicholas J. Bryan, Justin Salamon, Mark Cartwright, Juan Pablo Bello

摘要:

少样本学习的目标是训练模型,这些模型可以识别出只有少量标记示例(称为支持集)的新类别。虽然该领域近年来取得了显着的进展,但他们通常集中在多类图像分类上。相比之下,音频由于声音重叠而通常是多标签的,从而导致诸如复调和信噪比(SNR)之类的独特属性。这导致了关于这样的音频属性可能对少样本学习系统设计、性能和人机交互具有的影响的未回答的问题,因为通常由用户来收集和提供推理时间支持集示例。我们通过一系列旨在阐明这些问题的答案的实验来解决这些问题。我们介绍了两个新的数据集,FSD-MIX-CLIPS和FSD-MIX-SED,其程序生成允许我们系统地探索这些问题。我们的实验导致了对少样本学习的音频特定见解,其中一些与图像领域的最新发现不一致:没有最好的一刀切的模型,方法和支持集选择标准。相反,它取决于预期的应用场景。我们的代码和数据可在https://github.com/wangyu/rethink-audio-fsl上获得。

英文摘要:

Few-shot learning aims to train models that can recognize novel classes given just a handful of labeled examples, known as the support set. While the field has seen notable advances in recent years, they have often focused on multi-class image classification. Audio, in contrast, is often multi-label due to overlapping sounds, resulting in unique properties such as polyphony and signal-to-noise ratios (SNR). This leads to unanswered questions concerning the impact such audio properties may have on few-shot learning system design, performance, and human-computer interaction, as it is typically up to the user to collect and provide inference-time support set examples. We address these questions through a series of experiments designed to elucidate the answers to these questions. We introduce two novel datasets, FSD-MIX-CLIPS and FSD-MIX-SED, whose programmatic generation allows us to explore these questions systematically. Our experiments lead to audio-specific insights on few-shot learning, some of which are at odds with recent findings in the image domain: there is no best one-size-fits-all model, method, and support set selection criterion. Rather, it depends on the expected application scenario. Our code and data are available at https://github.com/wangyu/rethink-audio-fsl.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑