List of useful data augmentation resources. You will find here some not common techniques, libraries, links to GitHub repos, papers, and others.

Pitch-shift audio clips quickly with PyTorch (CUDA supported)! Additional utilities for searching efficient transformations are included.

Time-stretch audio clips quickly with PyTorch (CUDA supported)! Additional utilities for searching efficient transformations are included.

⚡ Blazing fast audio augmentation in Python, powered by GPU for high-efficiency processing in machine learning and audio analysis tasks.

A tool/script for batch speech data enhancement with speed/volume/RIRS/MUSAN

A python library for generating different permutations of audible segments from audio files.

A ready-to-use pytorch dataloader for audio classification, speech classification, speaker recognition, etc. with in-GPU augmentations

Python augmentation toolkit for Automatic Music Transcription datasets

Implementation of audio, image, and spectrogram augmentation techniques provided by the librosa, Keras and audiomentations

Audio data loading and augmentations in JAX

End-to-end pipeline for training a custom keyword detection model with TensorFlow & TFLite expor

This repository contains the code and methodology used for the BirdCLEF 2024 Kaggle competition, where I achieved a rank of 55th out of 974 participants, earning a bronze medal. The goal of this competition was to build a model that can accurately classify bird sounds.

Train custom wake word models with openWakeWord. A granular 13-step pipeline with compatibility patches for torchaudio 2.10+, Piper TTS, and speechbrain. Generates tiny ONNX models (~200 KB) for real-time keyword detection — like building your own "Hey Siri" trigger. WSL2/Linux + CUDA required.

your new pocket-sized audio wizard

SoundScaper is an audio augmented reality mobile application that allows users to author, save and reload virtual, and spatially interactive, three-dimensional binaural soundscapes within physical, real world spaces.

Converting text to audio and applying audio augmentation

A Convolutional Neural Network that distinguishes between the speakers emotions. Comes with multiple preprocessors to improve the models performance.