首页|期刊导航|Zoological Research(Diversity and Conservation)|Frequency reflects region in birdsong recognition:Quantified by mutual information and mitigated through adaptive normalization

Frequency reflects region in birdsong recognition:Quantified by mutual information and mitigated through adaptive normalizationOA

中文摘要

Geographic and environmental heterogeneity generates pronounced variability in bird vocal dialects,complicating reliable species identification across spatially distinct populations.To quantify such dialectal divergence,maximum mean discrepancy(MMD)analysis was applied to multiple regional birdsong datasets,revealing significant distributional differences between recording locations.To overcome these challenges,an adaptive normalization and recognition framework was developed that integrates taskdriven feature normalization with a multi-head attention ResNet(MHAResNet)classifier.The normalization module dynamically reweights frequency,time,and channel dimensions according to their contextual importance,enhancing feature alignment across domains.The classifier concurrently distinguishes both species identity and regional provenance.This coupled architecture suppresses domain-induced variability while preserving salient acoustic cues critical for recognition.To dissect the contribution of individual feature dimensions,mutual information neural estimation(MINE)was employed to quantify their relevance to species and region classification.Across three geographically diverse birdsong datasets,the proposed method improved species recognition by an average of 2.9%and region recognition by 3.0%relative to non-normalized baselines.MINE results indicated that frequency features were the most predictive of geographic origin,whereas channel-based features most strongly encoded species discrimination.These results advance understanding of birdsong feature attribution and offer a scalable framework for acoustic biodiversity assessment across biogeographic gradients.

Zhu-Lin Hao;Meng-Kun Zhu;Jiang-Jian Xie;Chang-Qing Ding

School of Technology,Beijing Forestry University,Beijing 100083,ChinaSchool of Technology,Beijing Forestry University,Beijing 100083,ChinaSchool of Technology,Beijing Forestry University,Beijing 100083,China State Key Laboratory of Efficient Production of Forest Resources,Beijing Forestry University,Beijing 100083,China Research Center for Biodiversity Intelligent Monitoring,Beijing Forestry University,Beijing 100083,ChinaSchool of Ecology and Nature Conservation,Beijing Forestry University,Beijing 100083,China

生物科学

Birdsong recognitionDialectal variationMutual information neural estimationMaximum mean discrepancy analysisAcoustic feature normalization

《Zoological Research(Diversity and Conservation)》 2025 (3)

P.196-206,11

supported by the National Natural Science Foundation of China(62303063,32270554)National Key Research and Development Program of China(2024YFF1307202)Beijing Natural Science Foundation(5252014)Beijing Forestry University Science and Technology Innovation Program(2024XY-G002).

10.24272/j.issn.2097-3772.2025.015

评论