Medical student and aspiring computational biomedicine researcher. Current and former World Record Holder of the Euler–Mascheroni Constant, Apéry’s Constant, Catalan’s Constant, and other mathematical constants as of May 2023. If you are interested in receiving assistance to set a world record for various mathematical constants, including Pi, contact me using the email below. Copying or using any website content without permission is prohibited.

2 minute read

Repository

https://github.com/tensorflow/magenta

Introduction

Linkin Park fans cannot forget their music. Sadly, we cannot hear it again, because of the death of the late Chester Bennington. Although replicating his voice is unfortunately impossible in 2019 based on current technology, replicating all the other instruments is possible.

I have tried using two NVIDIA Tesla V100 16GB SXM2 GPUs and the MusicVAE model (https://magenta.tensorflow.org/music-vae) to generate music based on based on machine learning and MIDI files of songs from the albums of Linkin Park.

Post Body

I became a fan of Linkin Park a bit before his death. That was when I was starting to listen to a lot of music. I didn’t before. And it was shocking to see him go.

Because of this, when I found out about the Magenta project while I was working on a programming project related to music, I thought instantly of Linkin Park. If I could replicate their nu-metal music using Magenta.

I wasted no time and immediately trained around 15 Linkin Park Midi files, composed of songs from the nu-metal albums Hybrid Theory, and Meteora. This Link is the downloadable .tfrecord file generated using This Manual.

Refer to Training Your Own MusicVAE to learn how to train your own model. I have used the hierdec-trio_16bar config and you will want to use the same if you are working on replicating popular music or band music like me. You can also generate piano music like Bach with the Performance RNN Model in the same package.

It took me about a week on one Tesla V100 GPU for 200k steps, in a batch size of 32.

Generation was done using This Command, and five samples were generated.

image.png image.png image.png image.png image.png

The results were not impressive. Due to overfitting, the sounds resembled one specific song from one of Linkin Park’s nu-metal albums. The results are available Here. Can you guess which song they resemble?

In my next post, I will address the expansion of data trained to create actual original music that replicates Linkin Park. This one is gonna be actually good.

Although I think my first test was unsuccessful, please rate the generated music and let me know what you think!

If you want to generate new music using this trained model, Download the Checkpoint from Here and use Tensorflow Magenta to generate using This Command.

Resources

MusicVAE model (https://magenta.tensorflow.org/music-vae) in Tensorflow Magenta (https://github.com/tensorflow/magenta)
FreeMidi.org (https://freemidi.org/)
FLAC Synthesizer by MuseScore (https://musescore.com or https://github.com/musescore/MuseScore)

This is the first post of a series of blog posts on machine-generated music. The collection of posts will expand to cover new genres, music artists, or bands.

Second Post

You May Also Enjoy

Breaking the Pi World Record?

9 minute read

Here I make a qualitative cost/time estimate on the feasibility and what is required for computing the next world record for Pi and the necessary other steps...