pamseek

pamseek is an underwater Passive Acoustic Monitoring (PAM) data analysis package designed to process large datasets using parallel processing. It streamlines and standardizes the processing, analysis, and reporting of underwater soundscape metrics in a fast and reproducible way.

This package leverages Python's multiprocessing module to efficiently handle large audio datasets by utilizing multiple CPU cores. This parallel processing capability significantly speeds up PAM data analysis, making it ideal for high-throughput workflows and long-term monitoring projects.

Features

• Parallel Audio Processing: Efficiently load and process multiple audio files simultaneously using multicore processing
• Spectrogram Visualization: Generate high-quality time-frequency representations of acoustic data
• LTSA Analysis: Calculate and visualize Long-Term Spectral Averages for extended monitoring periods
• PSD Computation: Perform Power Spectral Density analysis for detailed frequency domain assessment
• SPL Measurement: Calculate Sound Pressure Level metrics to quantify underwater sound intensity
• Broadband and Band-Limited Analysis: Conduct 1/3 octave band and other filter-based analyses
• Signal Detection: Integrate whistle detection functionality from soundscape_IR
• Annotation Tools: Process and modify annotations created with Raven Software
• Advanced Processing: Implement hydrophone calibration, timestamp extraction, and various filtering techniques (Butterworth, Chebyshev, frequency domain)
• Future Development: Upcoming CNN classification capabilities based on opensoundscape

Installation

Install pamseek using pip:

pip install pamseek

Quick Start

import pamseek

# Process audio files using multiple cores
DATA_PATH = r"E:\Hydrophone\2024-10"
ds_toctave_125Hz = pamseek.process_audio_files(
    path=DATA_PATH,
    sensitivity=-170.4,
    gain=2.05,
    fs=96000,
    window='hann',
    window_length=0.08533,
    overlap=0.5,
    scaling='density',
    low_f=111,
    high_f=140,
    n_processes=None,  # Uses all available cores by default
    output_dir="E:\Hydrophone\output",
    output_filename="OToctave_band_125Hz.nc"
)

Processing displays a progress bar and automatically skips corrupted files. Results are saved in NetCDF format with a confirmation message showing the file location.

Examples

LTSA (Long-Term Spectral Average)

pamseek.plot_ltsa(ds)

SPL and 1/3 Octave Band Analysis

Power Spectral Density (PSD)

Documentation

For detailed documentation and examples, visit our documentation site (coming soon).

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License.

Acknowledgments

• soundscape_IR for whistle detection algorithms
• opensoundscape for annotation tools and CNN classification framework
• Contributors and researchers from the underwater acoustics community