使用逐层相关性传播解释端到端深度学习语音声源定位模型
Interpreting End-to-End Deep Learning Models for Speech Source Localization Using Layer-wise Relevance Propagation
AI总结：
本文采用LRP技术分析两个端到端语音声源定位深度学习模型，发现其通过降噪和去混响处理输入信号以计算更准确的统计相关性，并利用GCC-PHAT验证了相关性信号能获得更准确的时延估计。
摘要：
深度学习模型在信号处理领域被广泛应用，然而其内部工作过程通常被视为黑盒。本文研究将可解释人工智能（XAI）技术应用于基于学习的端到端语音声源定位模型。我们考虑逐层相关性传播（LRP）技术，该技术旨在确定输入的哪些部分对输出预测更为重要。使用LRP，我们分析了两个最先进的模型，它们具有不同的架构复杂度，将麦克风采集的音频信号映射到声源的笛卡尔坐标。具体而言，我们检查了与这两个模型输入特征相关的相关性，并发现这两个网络都对麦克风信号进行降噪和去混响处理，以计算它们之间更准确的统计相关性，从而定位声源。为了进一步证明这一事实，我们使用麦克风信号和从这两个网络提取的相关性信号，通过带相位变换的广义互相关（GCC-PHAT）估计到达时间差（TDoAs），并表明通过后者我们获得了更准确的时延估计结果。
英文摘要：
Deep learning models are widely applied in the signal processing community, yet their inner working procedure is often treated as a black box. In this paper, we investigate the use of eXplainable Artificial Intelligence (XAI) techniques to learning-based end-to-end speech source localization models. We consider the Layer-wise Relevance Propagation (LRP) technique, which aims to determine which parts of the input are more important for the output prediction. Using LRP we analyze two state-of-the-art models, of differing architectural complexity that map audio signals acquired by the microphones to the cartesian coordinates of the source. Specifically, we inspect the relevance associated with the input features of the two models and discover that both networks denoise and de-reverberate the microphone signals to compute more accurate statistical correlations between them and consequently localize the sources. To further demonstrate this fact, we estimate the Time-Difference of Arrivals (TDoAs) via the Generalized Cross Correlation with Phase Transform (GCC-PHAT) using both microphone signals and relevance signals extracted from the two networks and show that through the latter we obtain more accurate time-delay estimation results.
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。
