A machine learning application in music has emerged. This is Lo-fi Player.
A web application allowing you to produce your own music with the help of the artificial intelligence was born. This is Lo-Fi Player created by research team Google Magenta.
Lo-Fi Player: Become a Lo-Fi Music Composer Using Artificial Intelligence Click to tweet
In this section:
Lo-Fi Player Concept
Lo-Fi Player is an opensource application that presents itself as a game.
We see a person in a room filled with objects. You will have to interact with each object to influence music basic.
For example, if you click on the television, you will have access to an interface allowing you to modify the melody which will merge with the current one.
We won't tell you more so as not to spoil the surprise.
You can go to lofi player website to try the experiment. It is even possible to share your Lo-Fi composition with your friends.
Archive – Youtube version
A youtube version of the project currently exists. It allows all participants to change the music via the chat room in a collaborative way.
An artificial intelligence will apply the commands randomly. The funny thing is that every time, the changes do not spoil the rhythm of the resulting music.
Changes and melodies seamlessly merge with the music being played.
The order list can be found in the description part of the youtube page of the Lo-fi player.
We talk a lot about Lo-Fi music but in reality what is it exactly?
What is Lo-Fi music?
Lo-Fi (Low Fidelity) is a term used to refer to music or productions considered to be of poor quality or likely to provide low fidelity.
The imperfections we are talking about can be inaudible sounds during production made audible (intentionally or not).
These sounds can come from theenvironment in which the work is produced as noises of Car or surrounding people.
La Lo-Fi music became very popular thanks to youtube and the covid-19 pandemic.
Author's Note:
I discovered the Lo-Fi music a few days before the Lo-fi Player application and I must say that it is very quickly captivating.
This genre, which is not really a genre, is ideal for programming and editorial work. It's definitely my second type of music after anime osts... .
Who are the creators of the Lo-Fi Player project?
Vibert Thio is a Summer 2020 Magenta Intern working with Researcher Douglas Eck. The vision and execution of this project belong 100% to Vibert!
Doug co-wrote the blog,. He also helped manage the project and, along with other team members, had many in-depth design and technical conversations with Vibert.
“Due to COVID-19, this project was particularly difficult. We all worked from home and faced many time zones: Vibert worked in Taipei, Doug from Paris and many Magentans from California. Most of the Magenta team still haven't met Vibert in person! Under these circumstances, a shared YouTube music streamer that bridges the physical distance, seems like an appropriate outcome.”
Animated Sunset GIF was created by amazing 2D artist / animator Sheena Tiong and is used with her permission. All other artwork was created by Vibert.
From Vibert:
“Thank you to the beautiful people at Magenta for helping me realize this project including Fjord Hawthorne, Andy Coenen, Monica Dinculescu and others. Thanks also to Damien Henry at Google Arts and Culture and Amit Pitaru and others at Google Creative Lab New York. Thanks to Conehead and EVE (aka 酷酷小乖乖) for the music tips. Thank you very much to everyone who gave me feedback. »
source: magenta.tensorflow.org/lofi-player
The project is based on the Magenta machine learning framework. Developers IT could play with this artificial intelligence tool.
Conclusion
That's all for this article, you may be interested in FL Studio's open source alternative et Darwin Mouse: The computer that imitates the brain
