Mixture-Constrained Max Pooling Improves Separation-Based Bird Species Classification

2026-07-03

AI Summary

This study addresses the challenges posed by overlapping bird vocalizations and incomplete labels in field recordings, which severely degrade multi-species classification performance. To mitigate these issues, the authors propose a source separation–based preprocessing framework that integrates two separation models—FTRNN and TF-Locoformer—and introduces a novel Mixed-Constrained Max-pooling (MCM) mechanism. MCM leverages species presence priors derived from the original mixture to constrain predictions from individual separation channels, thereby effectively suppressing false positives caused by separation errors. Experimental results on two real-world datasets demonstrate that the proposed approach consistently outperforms both single-separator baselines and standard max-pooling strategies across multiple evaluation metrics, confirming the efficacy of combining source separation with MCM for enhancing true positives and reducing false positives.

Abstract

Bird species classification from field recordings remains challenging due to overlapping vocalizations and incomplete species labels. We study source separation as a preprocessing for bird species classification to improve multi-species detection. Specifically, we employ an ensemble of two separators, FTRNN and TF-Locoformer, both trained with mixture invariant training (MixIT). To address the false positive gain caused by separation errors in separated outputs, we propose mixture-constrained max pooling (MCM), which clips the predicted probability from each separated channel based on the corresponding species probability in the original mixture. The classifier is applied to each separated output and the original mixture independently, and MCM aggregates the predictions into a final per-species probability. Experiments on two real-world datasets show that the ensemble outperforms individual separators and MCM outperforms standard max pooling across multiple metrics, and reveal that separation leads to both true positive gain for present species and false positive gain for absent species.

Problem

• bird species classification
• overlapping vocalizations
• incomplete species labels
• source separation

Innovation

• mixture-constrained max pooling
• source separation
• bird species classification
• MixIT
• ensemble separation