Hi Hi Hi,
Di postingan ini gw akan nge-share artikel yang dibuat oleh salah seorang mahasiswa/i gw mengenai Artificial Intelligence. Postingan ini salah satu bentuk apresiasi gw atas pemikiran mereka. Nama pembuat ada di judul postingan.
There are so many application of artificial intelligence including medial diagnosis, eletronic trading, speech recognition, face recognition and so many more. But the artifical intelligence and music (AIM) is less known by people, but it’s already published long ago by a Russian researcher R.Kh.Zaripov in a worldwide paper on algorithmic music composing using the “Ural-1” computer and also at the 1965 an invetor named Ray Kurzweil show a piano piece that create by a computer that was capable of pattern recognition in various compositions, the computer was able to analyze and use the patterns to create melodies. And at the 2014 there is a company named Amper Music, that using artificial intelligence to compose music and customize music and beside Amper Music, there is also Jukedeck, Orb Composer, AIVA, Melodrive, and so many more music composer that based on Artificial Intelligence.
These AI-based music composer mostly work by using the Neural Network and Deep Learning method, or we can say that we give the AI-based music composer a tons of sourcefull material, like pop, disco, classic, and so many others kind of music, which the music composer software will learn the all of the resource the inputted to and analyze to bring out a pattern. The pattern is in form of like tempo, length, chords, how high is a note should be, and from all of that the music composer makes its own music. But there is difference between the platforms, there is some that purely hard-coded based on music theory to give the output and some purely by examining data (deep learning). However, there is a same thing that all of these AI-based music composer have, it is that the music they composed is making less senses every time we hear it again, because the music was composed based on collection of beat, melodies, chord, and tempo that already exist, not based on a feeling that human got, so it’s really not perfected yet.
There have been several music tracks that composed with artificial intelligence music composer, such as the “Break Free” that created by Taryn Southern at the 2017. She is using the Amper Music to make the music composition , but she also need to change a bit of the composition because after hearing the “raw result” from the Amper Music it is really flat to be called as a pop music, so she changed the key, switched the instrument, and finally writing and perfoming the vocal to complete the song “Break Free”. The “raw result” might be incomplete, but it’s still sound quite satisfying for a music that composed by a software.
The most perfected and complex AI-based music compose might be AIVA, AIVA is An AI-based music composer that became the world very first AI-based music composer that recognized by the SACEM (Société des auteurs, compositeurs et éditeurs de musique) a French professional association collecting payment of artist rights and distributing the rights to the original songwriters, composers, and music publisher. The music that AIVA composes is classic music, there is a track the AIVA create, it is named “Genesis, Symphonic Fantasy in A Minor, Op.21”, the sound of the track is very unique and also not so flat, it is just feel like the track was composed by a human, written and felt by the human itself.
Maybe AI-based music composer at this age is not really perfected yet, because it cannot compose music for all kind of music, especially for the pop music, oftenly the music the software compose is too flat to be heard. But for the classic music, AIVA already give the very satisfying result, this means at some point in the future, AI-based will working like a human music composer, composing the perfect music for a genre that the customer want. With that, it means the job of music composer is at some point is at risks at the future, because the AI-based music composer is some way cheaper and more efficient in time, so this work very well for a company that running on low budget and short of time when running a project.
Nice article Halid. Keep the good work! Check his linkedin in https://www.linkedin.com/in/halid-wi-1b0257164/
See you di postingan selanjutnya.