Double Yu Expo: Neural network for making music

2024: Neural Network Launch

In July 2024, an open neural network for creating music was launched in Russia, developed by engineers of the Perm company Double Yu Expo. The new technology based on artificial intelligence is able to generate musical works in modern genres, create unique lyrics and customize compositions to suit the tastes of the target audience.

According to Ekaterina Kokorina, executive director of Double Yu Expo, the neural network greatly simplifies and speeds up the process of creating musical compositions, helping to analyze data and create original works.

Unsplash

It inspires new musical directions. At the moment, we are still choosing a business model that can monetize this technology, after which we will be able to begin cooperation with artists and composers, - said Ekaterina Kokorina.

To develop an innovative solution, engineers used the Python programming language used for data processing and machine learning. In addition, machine learning libraries including TensorFlow and PyTorch were deployed to train the neural network. The developers also provided a RESTful API for interacting with client applications.

A feature of the technology is the ability to create not only an instrumental, but also a vocal part of the work. This was made possible by the use of extensive data libraries based on thousands of popular works from leading music services, as well as a unique array of dialogue databases of real service robots collected during communication with people around the world.

Double Yu Expo already has experience in the development of innovative technologies. She previously introduced a humanoid named robot Ardi. Then, a robot created on the basis Robo-C of the company's robotic platform Promobot attracted the attention of about 10 thousand people.^[1]

Notes