animal-sounds

Chunk wav

Chunk_wav component splits .wav files into smaller chunks. This is a step in the preparation process for deep learning models.

Table of Contents

• How does it work
• Software dependencies
• Usage
  • Shell script
• Remarks

How does it work

Here you can find the inputs to make_chunks.py:

• Directory of wav files
• Desired length of wav files in seconds
• Overlap of two following chunks in seconds
• Directory of output wav files

By default, desired length of wav files is 0.5 seconds. There is an overlap of 0.25 seconds between each of two following chunks.

Software requirements

• Python 3
• numpy ~1.23.2
• scipy~1.9.1

Basic usage

This folder contains a python script to process the original .wav files and a .sh shell script to call the python script. The python script can be run from command line.

python make_chunks.py 
        --input_dir
        --chunk_length 
        --overlap
        --output_dir

Shell script

Here is a convenient shell script that runs the python script for the entire dataset.

$ ./split_dataset.sh 

Remarks

This site is open source. Improve this page.