Audio data augmentation tool for machine learning projects
Project description
Audio-Augment
Audio data augmentation tool for machine learning projects
This library augments audio training datasets by transforming provided WAV files in 10 different ways:
- Waveform Inversion
- Highpass Filter
- Lowpass Filter
- Bandpass Filter
- Add noise (normal, uniform)
- Pitch shift (low, high)
- Time shift (slow, fast)
Directory Setup
Create two folders in your current working directory, one called unprocessed
and another called processed
.
Place all audio samples that you want to transform in the unprocessed folder (must be WAV format).
Run this method and it will transform all audio files in your unprocessed folder,
and save the new set of augmented samples in the processed folder.
Requirements
-
numpy
-
pandas
-
matplotlib
-
soundfile
-
librosa
For questions about this library, email wesleylaurencetech@gmail.com
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
AudioAugment-0.0.3.tar.gz
(5.7 kB
view hashes)
Built Distribution
Close
Hashes for AudioAugment-0.0.3-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | df5b05a00891384a1d17d4f89ccbd28a2bd3ba138a85a639f9f68089b6470c1f |
|
MD5 | 28193f04adb876a4b8964e99f1f45246 |
|
BLAKE2b-256 | b0b4e4c02e9c99554fb030bcfe6fea03e16bc9095c4b37c34e8fe8e426f8273d |