Species Identification from Bioacoustics

Solution for BirdCLEF+ 2025

Code and experiments for my submission to the BirdCLEF+ 2025 Kaggle competition.

Demonstrating the "Reading the Robot Mind" system using a bird audio recognition AI system.

基于 BirdCLEF 2026 的声音鸟类识别课程项目，包含语音特征工程、传统机器学习、CNN/CRNN 对比、GroupKFold 泛化分析与 Gradio 可视化演示。

CPU-ready bioacoustic species recognition pipeline for Kaggle BirdCLEF+ 2026.

Bird sound recognition & language system — identify species and call types from audio (CLI + Gradio GUI)

Design of an ablation study for a Machine Learning pipeline. The effect of preprocessing, model or postprocessing modules can be automatically tested.

Animal sound classifier for BirdCLEF+ 2025 — EfficientNet B0 + FastAPI, with a Gemini agent that checks whether detections are plausible given your context.

I made this project with 2 fellow students as a solution for the 2024 BirdCLEF competition.

BirdCLEF+ 2025 multi-label bird sound event detection — EfficientNet-B0 SED + attention pooling (val AUC 0.8529)

Bioacoustic audio classification and soundscape analysis using deep learning and ensemble methods.

Ablation study for BirdCLEF+ 2026 — companion code for Kaggle notebook series and CLEF 2026 working note

BirdCLEF+ 2026 bioacoustic detection with Perch embeddings, ProtoSSM, OOF stacking and TTA

Acoustic species ID on BirdCLEF 2026 — BirdNET embeddings + ML vs. from-scratch CNN, testing AudioLDM2 synthetic-data enrichment (0.894 macro ROC-AUC).