InterBias-SV: Compound Conditions in Speaker Verification

2026-09-28

AI Summary

This study addresses the challenge of additively evaluating compound condition effects, such as noise and channel distortion, in speaker verification. We propose a joint error versus marginal error comparison paradigm and construct a quaternary contrastive framework. Through multi-corpus benchmarking and ten-million-scale trial matching analysis, this work systematically reveals misconceptions regarding saturation mechanisms under near-random equal error rate (EER) conditions and the instability of demographic disparity ratios. The project yields a benchmark dataset comprising 4,068 records, quantifies average EER discrepancies across encoders, and provides a fully reproducible toolchain with clearly defined experimental boundaries. These contributions establish a rigorous empirical foundation for assessing model robustness under compound conditions in speaker verification systems.

Abstract

Speaker verification systems encounter combinations of noise, channel distortion, and changes in speech. Evaluating each condition separately does not establish whether their effects add. InterBias-SV organises this question around a four-term comparison: joint error, two marginal errors, and a common reference. Its results artefact contains 4,068 scored records across 17 experiments, 12 encoder labels, and six speech corpora, totalling 12 million trial evaluations. Three experiment families contain the same-corpus terms needed to compute additive contrasts. For labels assigned to speaker-trained encoders, their mean contrasts are +0.0026, +0.0088, and +0.0024 in equal error rate (EER), with larger variation across settings. These descriptive averages do not establish equivalence to additivity: trial matching, checkpoint identity, and parts of the condition metadata remain unverified. We also examine two interpretation problems. Near-chance EER can make additive predictions difficult to interpret, but chance performance is not a hard EER ceiling, and correlation with the prediction does not identify a saturation mechanism. Ratios of demographic gaps are unstable when their clean reference is near zero; absolute gaps provide a more direct summary. The benchmark provides condition definitions, analysis scripts, and explicit requirements for interpretable compound-condition comparisons, while separating recomputable summaries from claims that require further experimental validation.

Problem

• Speaker Verification
• Compound Conditions
• Additivity
• Equal Error Rate
• Demographic Gaps

Innovation

• Speaker Verification
• Compound Conditions
• Additive Interaction
• Benchmark
• Equal Error Rate