AudioCarver

Seam Carving as a Data Augmentation Technique

For more detail please check out my website

Audio Processing Library

A Python library for performing seam carving on audio files, converting audio to matrix and back, and generating spectrograms.

Table of Contents

• Installation
• Usage
• Features
• Examples
• Documentation
• License
• Contact

Installation

You can install the library using pip:

pip install audio_carver

Usage

Example in a python file:

from audio_carver import carve_audio, sig_to_wav

# Carving in time domain and writing to output
magnitude, phase = carve_audio(number_of_seams, magnitude, phase, is_vertical=True) # default true
sig_to_wav('output.wav', magnitude, phase)

Example in terminal(Recommended):

$ python main.py <input_file.wav> <number_of_seams> <--carve_time> # remove to carve in frequency

Features

Current Supported Audio Formats: .wav

Audio Processing Capabilities: Spectral analysis, Time-domain operations, Signal modification

Input Requirements: None

Output: Mono signal with sampling rate inherited from the original input file's sampling rate

VIsualization: Spectrogram comparisons of before/after carving

Examples

How to load an audio file:

from audio_carver import *

audio, sampling_rate = load_wavfile('filename.wav')

Find the minimum seam:

from audio_carver import *

minimum_energy_seam = min_vertical_seam_energy(magnitude)

Documentation

Documentaion will be uploaded soon.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contact

Please contact alicjam@uw.edu for questions, suggestions, etc. Thank You!