This weeks blog post is about a well known (and somewhat infamous) Japanese music synthesizer software known as Vocaloid. It was originally developed in 2000 by Kenmochi Hideki at the Pompeu Fabra University in Barcelona, Spain. The project, which was initially not intended for a widespread commercial release, was backed by the Yamaha Corporation. The software allows users to create songs and synthesize singing by typing in lyrics and a melody. The voices samples were provided by different actors and singers, and there are specific “characters” that have their own unique voice. The software has allowed for the creation of many different songs and projects by both independent and professional artists and composers, like Hachi, DECO27 and sasakure.UK to name a few.
How the software works is that you use the provided voice samples and libraries of sound and run them through the synthesizer. Here is a visual diagram of how the sounds and program work together.
What this is really is that the voice samples are joined together in a sequence according to the lyrics and melody provided. To avoid clunky, disjointed singing, the vocaloid synthesizer tries to smooth out the sound waves as best it can to recreate a person singing. Vocaloid songs almost always sound robotic in nature however.
The vocaloid characters themselves have built up something of a fanbase on the internet, but for the sake of keeping this blog post short I won’t really go too much into that. Although there are vocaloids for other languages besides Japanese (like Korean, English and Spanish), the Japanese vocaloids tend to be used the most and the most popular.
Here is an example of a Vocaloid singing a song (in Japanese).
Vocaloid has inspired many creators (animators, artists, composers etc.) and a lot of songs usually have their own fanmade animated music videos (again in Japanese, you don’t have to watch the following is just provided as an example)