iver56/torch-audiomentations
Fast audio data augmentation in PyTorch. Inspired by audiomentations. Useful for deep learning.
This tool helps machine learning engineers and researchers prepare audio datasets for training deep learning models. It takes raw audio recordings, either mono or multi-channel, and applies various realistic modifications like adding background noise, altering pitch, or adjusting volume. The output is augmented audio data that helps models learn more robustly from diverse sound environments.
1,136 stars.
Use this if you are a machine learning engineer working with audio data and need to quickly and efficiently generate varied training examples on a GPU to improve your model's performance.
Not ideal if you are working with non-audio data, require highly specialized or non-differentiable audio processing not included, or are not using PyTorch for your deep learning models.
Stars
1,136
Forks
100
Language
Python
License
MIT
Category
Last pushed
Nov 24, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/iver56/torch-audiomentations"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
descriptinc/descript-audio-codec
State-of-the-art audio codec with 90x compression factor. Supports 44.1kHz, 24kHz, and 16kHz...
drethage/speech-denoising-wavenet
A neural network for end-to-end speech denoising
YuanGongND/ast
Code for the Interspeech 2021 paper "AST: Audio Spectrogram Transformer".
lmnt-com/wavegrad
A fast, high-quality neural vocoder.
madhavmk/Noise2Noise-audio_denoising_without_clean_training_data
Source code for the paper titled "Speech Denoising without Clean Training Data: a Noise2Noise...