THE SIGNAL STATE GAME SOLUTIONS SERIES

You can encode these musical events-which have both a time and a pitch dimension-to a sequence of tokens.įigure 1. The soprano is green, the alto is orange, the tenor is blue, and the bass is red. Usually, all four voices sing at the same time but with different, harmonic pitches.įigure 1 visualizes the voices with pitch color coding. Soprano and alto are female voices, and tenor and bass are male voices. A chorale features four voices: soprano, alto, tenor, and bass. The example below is a four-part chorale in its piano roll representation. Although this hierarchy can be considered a tree, it is possible to encode everything as a linear sequence, making it an ideal representation for decoder-only language models. And each bar holds a sequence of note-on, time-delta, and note-off events. Each track consists of several bars (4, 8, or 16 bars, depending on the use case).

A piece of music consists of different tracks for different instruments: drums, guitars, bass, and piano, for example. This encoding represents pieces of music as a hierarchy. For this representation, we took inspiration from the mmmtrack encoding. The MIDI format is a non-human-readable representation of music which, in order to train a Causal Language Model, has to be mapped to a readable token representation. AI music composed using a GPT trained on the MetaMIDI Dataset MetaMIDI Dataset with 463K MIDI files, again of varying genres and styles.Lakh MIDI Dataset and its Clean subset (176K and 15K MIDI files respectively), with a mixed variety of genres and styles.JS Fake Chorales Dataset with 500 fake chorales in the style of J.S.These experiments used several sets of MIDI files, including: The MIDI standard has been designed to electronically store music information. In research and application, MIDI-files have proven to be fruitful sources for musical material. Usually, this requires tokens for which note starts (C, D, E, F, G), how much time passes (quarter notes or eighth notes, for example), and which note ends.

Instead, you map anything that resembles sheet music to a token representation. The second approach does not work with the pure audio. One approach involves training on the music represented as pure audio (WAV files or MP3s). There are two major classes when it comes to datasets for Computational Music. The DGX-2 boosted progress significantly in both data preprocessing and training language models.

THE SIGNAL STATE GAME SOLUTIONS SERIES

This post provides an account of a series of experiments performed in the field of AI music using the NVIDIA DGX-2. This implies the need for a log of GPU compute to train the models effectively for rapid development, prototyping, and iteration. Starting with 50GB of uncompressed text files for language generation is no surprise. Large datasets are required to use language models in these domains. The same architecture can be used for music composition.

Specifically, these models have been used as powerful tools for writing, programming, and painting. Language models such as the NVIDIA Megatron-LM and OpenAI GPT-2 and GPT-3 have been used to enhance human productivity and creativity.

The signal state game solutions