Music and Audio

Audio transforms for FastAI

Instructional notebooks on music information retrieval.

Get media information from YouTube and Vimeo videos, Spotify tracks and Discogs releases.

ScorePerformer: Expressive Piano Performance Rendering with Fine-Grained Control (ISMIR 2023)

Official implementation of "Contrastive Audio-Language Learning for Music" (ISMIR 2022)

List of academic resources on Multimodal ML for Music

Structured Data from Ishkur's Guide to Electronic Music. Working Mirror for v2.5 here: https://igorbrigadir.github.io/ishkurs-guide-dataset/

This repository contains a web application that integrates with a music recommendation system, which leverages a dataset of 3,415 audio files, each lasting thirty seconds, utilising a Locality-Sensitive Hashing (LSH) implementation to determine rhythmic similarity, as part of an assignment for the Fundamental of Big Data Analytics (DS2004) course.

[REPO] Create a music genre dataset (metadata or audio) for classification using the Spotify Web API

Open source library that allows you to efficiently compute audio features.

A CNN which converts piano audio to a simplified MIDI format

USB Audio DAC using inexpensive STM32F411 / 401 "Black Pill" and PCM5102A modules

AudioLDM training, finetuning, evaluation and inference.

FPGA based USB 2.0 high speed audio interface featuring multiple optical ADAT inputs and outputs

Code for Vision-Infused Deep Audio Inpainting (ICCV 2019)

Basics of Musical Instruments Classification using Machine Learning

Audio Coding Notebooks and Tutorials

the dataset used in the paper https://drive.google.com/file/d/0B4bIMgQlCAuqdGVRbVNNbzJfeUU/view

Lyrics-to-audio-alignement system. Based on Machine Learning Algorithms: Hidden Markov Models with Viterbi forced alignment. The alignment is explicitly aware of durations of musical notes. The phonetic model are classified with MLP Deep Neural Network.

DALI: a large Dataset of synchronised Audio, LyrIcs and vocal notes.

Control mechanisms to the U-Net architecture for doing multiple source separation instruments

A Freesound Community Sampler

Creation of an embedding space using unsupervised triplet loss and Tile2Vec that can be used for a variety of downstream tasks

Neural network to classify some styles of Electronic music

This is the codes repository for the paper "Emotion-Guided Music Accompaniment Generation based on VAE".

The code for the MaD TwinNet. Demo page:

Classification wrappers for 4 Deep Learning Embedded Runtimes for Music Information Retrieval, plus utilities to convert a TensorFlow Model

Repository for the IEEE/ACM TASLP 2023 Paper "Zero-Note Samba".

Code for "Deep Learning Based EDM Subgenre Classification using Mel-Spectrogram and Tempogram Features" arXiv:2110.08862, 2021.