Similar to how fashion displays, ad campaigns, and even shopping experiences wouldn’t be the same without their own soundtrack, music is a continuous companion that aids in identity expression, social interaction, and the formation of communities.
This tie between the two can be seen in the way that we identify particular fashions, such as leather jackets, with rock fans or in the way that followers of Taylor Swift choose their concert attire based on one of the singer’s “eras.”
An artificial intelligence program that can create playlists depending on an individual’s appearance has been developed by a research group at LCI Barcelona, which has investigated this link. Known by the name Groovify, this system was unveiled in June of last year at Barcelona’s Sónar+D music and innovation event.
“The artificial intelligence expert professor at LCI suggested creating an algorithm. Our suggestion came from the fashion department to link our algorithm to a study of various fashion looks. It would recognize visual cues to distinguish between various clothing types,” Manetti continued.
After that, the software generates up to five distinct prompts, each of which produces a unique song and rhythm that correspond to the user’s photo.
In addition to playlist recommendations based on an individual’s style of dressing, the outcome involves developing a mobile application that can suggest music based on an individual’s appearance. Artificial intelligence is also used to create a brand-new song that is derived from a database of artists connected to the Sónar Festival. Alongside this unheard song would be a fictitious cover created with image-generating software.
A thorough analysis of databases already in existence, covering styles, music, covers, and visual codes found in discographies and performance and audience photos, forms the foundation for the process of linking style and music.
“We wanted to go beyond simply generating suggestions based on clothing style; we sought to create an unreleased track generated by artificial intelligence and pay homage to vinyl with an imaginary cover,” Manetti told FashionUnited about the process.
Both the general audience and potential investors responded favorably to the concept during the Sónar Festival. The technology in question is a proprietary algorithm that has drawn more attention because of its uniqueness and exclusivity. The project’s creators intend to release version 2.0, an enhanced version, in October.
In order to increase the app’s relevance and utility at the market level, this new edition will broaden the database to incorporate other sources like Beatport, YouTube, and Spotify. The team is also getting ready to attend the Web Summit in Lisbon in November, where they plan to showcase the app’s beta version and seek collaborations and commercial prospects.
The LCI Barcelona fashion department managers Pedro Coelho, Estel Vilaseca, Anna Pallerols, David Carretero, and Salvatore Elefante support the project, which is headed by Alessandro Manetti and directed by Mariele Violano. Six students who work in graphic design, interior design, user support, setup, and breakdown are among the 19 members of the team.
