(Deep) House: Making AI-Generated House Music

By Medium - 2021-03-10

Description

People have been trying to make machine generated music for a long time. Some of the earliest examples were musicians punching holes in piano roles to create complex melodies unplayable by humans…

Summary

  • Making AI-Generated House Music Ever wonder what computers play when we’re not around?
  • The level 2 generated audio samples contain the long-range structure and semantics of the music but are low audio quality due to the compression.
  • Image by Dhariwal 2020) These upsamplers add in local musical structure like timbre and greatly improve the sound quality.
  • I ran SoundHound against all of the auto-generated tracks and got no matches, but it’s unclear if this is because Jukebox is overfitting vocal-removed songs which don’t create a match or if they’re genuinely novel samples that are just heavily inspired by the training dataset.

 

Topics

  1. Machine_Learning (0.23)
  2. Backend (0.23)
  3. NLP (0.17)

Similar Articles

LSTM for time series prediction

By KDnuggets - 2021-01-23

Learn how to develop a LSTM neural network with PyTorch on trading data to predict future prices by mimicking actual values of the time series data.

K-fold Cross Validation with PyTorch

By MachineCurve - 2021-02-02

Explanations and code examples showing you how to use K-fold Cross Validation for Machine Learning model evaluation/testing with PyTorch.