This model is designed to predict if birds are present in five-second clips of an audio file. The model's use is upstream of a larger modeling effort: to identify clips in weakly labeled audio files that contain bird sounds with high probability. The aim is not to perfectly predict bird presence (or have good calibration), but to strongly label a subset of audio clips in the Birdclef Kaggle competition datasets.
The 24 features are of the characteristics:
1. Sound frequency percentiles
https://www.tandfonline.com/doi/full/10.1080/09524622.2020.1730241
quantiles = [0.8,0.9,0.925,0.95,0.975,0.99,]
2. 13 Mel spectrogram cepstral coefficients
Summary statistics of zero crossing rates in 1-second segments
Mean, standard deviations, min, max
3. Zero crossing rate of the entire clip
These features summarize an entire clip, irrespective of position in waveform or spectrogram, and technically, the clip does not have to be 5 seconds long.
I built the model iteratively with some publicly a…
