Cross-Dialect Bird Species Recognition With Dialect-Calibrated Augmentation

MCML Authors

Björn Schuller

Prof. Dr.

Core PI

Health Informatics

Abstract

Dialect variation hampers automatic recognition of bird calls collected by passive acoustic monitoring (PAM). We address the problem on DB3V, a three-region, ten-species corpus of 8-s clips, and propose a deployable framework built on Time-Delay Neural Networks (TDNNs). Frequency-sensitive normalisation (Instance Frequency Normalisation and a gated Relaxed-IFN) is paired with gradient-reversal adversarial training to learn region-invariant embeddings. A multi-level augmentation scheme combines waveform perturbations, Mixup for rare classes, and CycleGAN transfer that synthesises Region 2 (Interior Plains)-style audio, with Dialect-Calibrated Augmentation (DCA) softly down-weighting synthetic samples to limit artifacts. Our system significantly outperforms general bioacoustic foundation models (e.g., BirdNET, Perch) in dialect sensitivity and lifts cross-dialect accuracy by up to twenty percentage points over baseline TDNNs. The model maintains robustness under real field noise from BirdCLEF soundscapes, while Grad-CAM and LIME analyses show that it concentrates on stable harmonic bands. The study demonstrates that lightweight, transparent, and dialect-resilient bird-sound recognition is attainable.

article DSA+26

---

IEEE Signal Processing Letters

33. Aug. 2026.

---

Authors

J. Ding • Q. Sun • A. Akman • B. W. Schuller

---

Links

DOI

---

Research Area

 B3 | Multimodal Perception