https://github.com/materialvision/augment_audio_tools

Simple python scripts to augment a dataset of audio
https://github.com/materialvision/augment_audio_tools

Last synced: 18 days ago
JSON representation

Simple python scripts to augment a dataset of audio

• Host: GitHub
• URL: https://github.com/materialvision/augment_audio_tools
• Owner: materialvision
• License: cc0-1.0
• Created: 2023-03-26T09:02:17.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-06-07T10:07:00.000Z (over 1 year ago)
• Last Synced: 2024-08-08T00:43:10.261Z (4 months ago)
• Language: Python
• Size: 17.6 KB
• Stars: 7
• Watchers: 1
• Forks: 1
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• project-awesome - materialvision/augment_audio_tools - Simple python scripts to augment a dataset of audio (Python)

README

        
# Audio Augmentation Tools for Machine Learning

This script provides a set of audio augmentations for machine learning purposes, in particular useful for the RAVE model https://github.com/acids-ircam/RAVE . It allows you to process audio files by changing their speed, resampling, splitting stereo files into mono, adding silence, and creating chunks.

## Features

- Change speed of the audio file

- Resample the audio file

- Split stereo files into mono

- Add silence to the audio file

- Create chunks of the audio file

## Requirements

- Python 3.6 or higher

- NumPy

- SoundFile

- Resampy

To install the required packages, you can run:

```bash

pip install numpy soundfile resampy

```

## Usage

To use this script, run it from the command line with the following arguments:

```

python augment_audio_speed.py   [--chunk_duration] [--split_stereo] [--add_silence] [--speed_change]

```

- `input_folder`: Path to the input folder containing audio files

- `output_folder`: Path to the output folder for processed files

- `--chunk_duration`: (optional) Duration of each chunk in seconds (default: 30 seconds)

- `--split_stereo`: (optional) Split stereo files into two mono files

- `--add_silence`: (optional) Length of silence in seconds added to the end of each sound file

- `--speed_change`: (optional) Speed change factor 0.0-0.9 (default: 0.0, no change)

Example:

```bash

python augment_audio_speed.py input_folder output_folder --chunk_duration 30 --split_stereo --add_silence 1.5 --speed_change 0.1

```

This will process all supported audio files in the `input_folder` and save the processed files to the `output_folder` with specified augmentations.

## Supported Audio Formats

The script supports the following audio file formats:

- .wav

- .flac

- .ogg

- .aiff

- .mp3