Audialab Emergent Drums

The WhatTheBeat tool enables users to understand the meanings behind their favorite songs by artists like Drake, Eminem, A.R. Rahman, Justin Bieber, Michael Jackson, and Taylor Swift, through the use of artificial intelligence.

DADABOTS is a fusion of a music group, a hackathon team, and a temporary research lab that merges music with mathematics. They employ deep learning algorithms such as Wavenets, NSynth, and SampleRNN to develop an AI capable of generating and replicating music. Their primary interest lies in math rock and black metal genres. They investigate techniques to model musical styles, produce music with specific timbres, and develop a legion of remix bots. They aim to expand the limits to sustain music's vitality and uncover fresh sounds.

MuseNet, developed by OpenAI, is a sophisticated neural network capable of creating 4-minute musical pieces using 10 different instruments and blending styles ranging from country to Mozart to the Beatles. It operates with the same versatile unsupervised technology as GPT-2, a vast transformer model designed to forecast the next token in a sequence, applicable to both audio and text. The model learns from MIDI file data and can produce samples in a selected style by beginning with a prompt. It utilizes multiple embeddings, including positional, timing, and structural embeddings, to provide the model with additional context.