PhiNet: Speaker Verification with Phonetic Interpretability
This addresses the need for interpretability in speaker verification for high-accountability applications, such as forensic analysis, though it is incremental in combining existing methods with phonetic features.
The paper tackled the lack of transparency in automatic speaker verification systems by proposing PhiNet, a network that enhances interpretability using phonetic evidence, achieving performance comparable to black-box models on datasets like VoxCeleb, SITW, and LibriSpeech.
Despite remarkable progress, automatic speaker verification (ASV) systems typically lack the transparency required for high-accountability applications. Motivated by how human experts perform forensic speaker comparison (FSC), we propose a speaker verification network with phonetic interpretability, PhiNet, designed to enhance both local and global interpretability by leveraging phonetic evidence in decision-making. For users, PhiNet provides detailed phonetic-level comparisons that enable manual inspection of speaker-specific features and facilitate a more critical evaluation of verification outcomes. For developers, it offers explicit reasoning behind verification decisions, simplifying error tracing and informing hyperparameter selection. In our experiments, we demonstrate PhiNet's interpretability with practical examples, including its application in analyzing the impact of different hyperparameters. We conduct both qualitative and quantitative evaluations of the proposed interpretability methods and assess speaker verification performance across multiple benchmark datasets, including VoxCeleb, SITW, and LibriSpeech. Results show that PhiNet achieves performance comparable to traditional black-box ASV models while offering meaningful, interpretable explanations for its decisions, bridging the gap between ASV and forensic analysis.