Neuron-level interpretations aim to explain network behaviors and properties by investigating neurons responsive to specific perceptual or structural input patterns. Although there is emerging work in the vision and language domains, none is explored for acoustic models. To bridge the gap, we introduce $\textit{AND}$, the first $\textbf{A}$udio $\textbf{N}$etwork $\textbf{D}$issection framework that automatically establishes natural language explanations of acoustic neurons based on highly-responsive audio. $\textit{AND}$ features the use of LLMs to summarize mutual acoustic features and identities among audio. Extensive experiments are conducted to verify $\textit{AND}$'s precise and informative descriptions. In addition, we demonstrate a potential use of $\textit{AND}$ for audio machine unlearning by conducting concept-specific pruning based on the generated descriptions. Finally, we highlight two acoustic model behaviors with analysis by $\textit{AND}$: (i) models discriminate audio with a combination of basic acoustic features rather than high-level abstract concepts; (ii) training strategies affect model behaviors and neuron interpretability -- supervised training guides neurons to gradually narrow their attention, while self-supervised learning encourages neurons to be polysemantic for exploring high-level features.

翻译：神经元层面的解释旨在通过研究对特定感知或结构输入模式敏感的神经元，来解释网络的行为与特性。尽管在视觉和语言领域已有新兴研究，但在声学模型方面尚未有探索。为填补这一空白，我们提出了 $\textit{AND}$，首个基于高响应音频自动建立声学神经元自然语言解释的 $\textbf{A}$udio $\textbf{N}$etwork $\textbf{D}$issection 框架。$\textit{AND}$ 的特点是利用大语言模型（LLM）来总结音频之间的共同声学特征与身份信息。我们进行了大量实验以验证 $\textit{AND}$ 所生成描述的准确性与信息量。此外，我们展示了 $\textit{AND}$ 在音频机器遗忘方面的一种潜在应用，即基于生成的描述进行特定概念剪枝。最后，我们通过 $\textit{AND}$ 的分析揭示了两种声学模型行为：（i）模型通过组合基本声学特征而非高层抽象概念来区分音频；（ii）训练策略影响模型行为与神经元可解释性——监督式训练引导神经元逐渐聚焦其注意力，而自监督学习则鼓励神经元成为多义词性以探索高层特征。