The concept and technology have attracted several business angels as well as Kima Ventures, the investment fund of French billionaire Xavier Niel, securing a total of €650.000 for AIVA.

To train the AI’s understanding of music, the team behind AIVA starts by feeding the algorithm with more than 15.000 scores of symphonic music, written by the best composers like Mozart, Beethoven and Bach. This means that AIVA knows every note and chord that’s played, including the pace and rhythm.

Using Deep Neural Networks, AIVA looks for patterns and rules in the scores to understand the basic style of the music. One way AIVA practises its understanding of music is by predicting what comes next in the track. Once it gets good at these predictions, it creates a set of mathematical rules for that style of music. Finally, AIVA is ready to compose totally unique scores.

Storytellers, film producers and the whole entertainment industry rely on music to turn moments into magic. But interactive content such as video games have hundreds of hours of gameplay, and typically only two hours of music, since human composers are limited by physical constraints. And rather than looping the same tunes over and over again, AI gives an alternative to augment human creative abilities, so that games may finally have hundreds of hours of original music.

Pierre Barreau, CEO of AIVA, comments: “Personalized music for games and individuals is what we are trying to achieve with AIVA. One of the path that this fundraising will help us achieve is conditional generation of music based on text, or ‘script-to-music’ translation, so that AIVA can compose a film or game soundtrack based directly on a script!”

However, the team behind AIVA – a team of musicians first and foremost – doesn’t believe that learning algorithms will replace artists. AI can fix writer’s block, and address use cases that composers alone cannot solve, but still need humans to be brought to life, with real instruments and a conductor’s irreplaceable artistic supervision.