Are you a song lover if yes then this invention is quite interesting for you. Researchers at the University of Waterloo, Canada, have recently developed a system for generating song lyrics that match the style of particular music artists. Let us know what is it and how does it work
**A system to generate new song lyrics that match the style of specific artists
Picture source credit: Tech xplore

This program not only generates lyrics but also generate in the way of your favorite artist. So for thus it uses a different set of data of different artist.

What are the components used in it?

It uses a variational autoencoder (VAE) with artist embeddings. VAE resembles a classical autoencoder. It is used to learn various data. It is a neural network consisting of an encoder, a decoder, and a loss function. It let us generate a model of data used for image generation

A CNN classifier trained to predict artists, which is used for analyzing images. CNNConvolutional Neural Network is a type of a computer-based algorithm to understand the features of the image and then match with the installed data. It uses MEL spectrograms which is a Mel-frequency cepstrum used in sound processing.of their artist song clips.

 Olga Vechtomova, a team member of this project is a music lover. He said Music is a passion of mine, and I was curious about whether a machine can generate lines that sound like the lyrics of my favorite artist.

 The lines generated from this software are

With a shadow beside,
All the tears inside I turned back to the earth.
She walks in a better moon
And I let my heart go.
You’ll be with freedom and shelter
Begin to the next mother earth.
And the light shines on the runway
Driving an endless ship on fire.
When I’m frightened in the air
Rooted in the light
There’s the garden in the darkness for us.
When the promise in the rain
And the stars they go.

 Also read: Technology that creates Poetry by your thoughts

This software is trained to generate any number of new diverse and coherent lyric lines. To generate the lines of specific artist a multi-dimensional vector of real numbers learned by a separate neural network. Then we set language of an artist to condition the generation of lyrics lines in the style of him.

This researched promises to generate powerful poetic lines and expressing novel thoughts not found in the original lyrics. In the future, the system created by Vechtomova and her colleagues could be used to inspire artists who are composing lyrics for new songs.

The researcher said this software does not make to replace the original artist instead of making them way to think a new direction. This software make their lyrics more powerful and more artistic. Vechtomova said. “In the music world, this could be analogous to a synthesizer that can generate an infinite number of sounds, from which an artist then creates a song.

 

Leave a Reply

Your email address will not be published. Required fields are marked *

*
*