Virtual Singer Uses Crowdsourced Songs To Become a Star In Japan (bloomberg.com)

An anonymous reader quotes Bloomberg. [Alternate version here]: During her 10-year career, she's released more than 100,000 songs in a variety of languages and opened shows for Lady Gaga. And yet Hatsune Miku, who boasts 2.5 million Facebook followers, doesn't actually exist -- at least not in the typical way we think of a flesh-and-blood diva. Miku is a computer-simulated pop star created more than a decade ago by Hiroyuki Ito, CEO of Crypton Future Media in Sapporo, Japan.

She started life as a piece of voice-synthesis software but since has evolved to become a singing sensation in her own right -- thanks to the creativity of her legions of fans. Crucial to Miku's success is the ability for devotees to purchase the Yamaha-powered Vocaloid software and write their own songs for the star to sing right back at them. Fans then can upload songs to the web and vie for the honor of having her perform them at "live" gigs, in which the computer-animated Miku takes center stage, surrounded by human guitarists, drummers and pianists.
Bloomberg's article includes some video clips of the virtual artist -- as well as her real-world fans.

Comments

Still.

[Templer421]

She isn't real.

Miku shows why openness actually promotes art

[DJ Rubbie]

Unlike typical Japanese media enterprises that exert their draconian copyright laws to squash usage of IP (including what Americans consider to be "fair use"), the creative forces that started Hatsune Miku put her design as part of the Creative Commons, thus freeing her design to amateur and professional artists alike for reuse. As a result, the original rights holder receive even greater recognition for their voice synthesizer software line from the artists creating all the derivative visual works involving her likeness.

Re:Well, it was only a matter of time

[AmiMoJo]

That's the thing about Hatsune Miku - she is a genuine grass roots effort, one of the least manufactured pop stars in Japan right now.

As TFA says, she started as a voice pack for the Vocaloid software synthesiser package. Fans used it to create songs and upload them to Nico Nico Douga, a site similar to YouTube. These were pure fan creations, the Vocaloid software was just the firm time that someone working from a PC at home could use high quality vocals without having to record them themselves.

As is often the case in Japan, other people started to supplement the popular songs with fan art and videos. In Japan it's common to have a mascot for your product, and since this was a female vocal synth they had already created Hatsne Miku with the intention of releasing other characters (voice packs) later.

Eventually independent record labels started collecting these songs and releasing them. Since the software is a just a synth like any other, the creators no more own the copyright than the creators of a drum synth or a real instrument would. As her popularity grew they started doing concerts and bigger CD releases, but most of the songs are still created by independent artists and posted for free online.

Re:Still (has problems with non-ortogarphic langs)

[Anonymous Coward]

> Japanese has some unusual properties that meant Vocaloid could sing in Japanese surprisingly well. The most important of these is that Japanese is basically a string of unconnected and discreet sounds.

Vocaloid was actually invented by a spanish science guy named Jordi Bonada while doing PhD in the USA. He met a japanese guest professor who had connections with the musical instruments branch of Yamaha zaibatsu and he convinced the board to pour a few million dollars into the research in exchange for IP rights. Luckily catalonian spanish dialect has the same "ortographical" property as the japanese language, so even today spanish remains the leading language for Vocaloid soundbanks alongside japanese. English is much rarer.

> So for an English voice synth when you type in "example", it has to run through a complex system that converts it into the vocal sounds for that word, before it even starts to consider adding expression

Yeah, a japanese language recording session for a Vocaloid avatar literally takes one afternoon and another for recap. An english soundbank takes about 6 weeks of microphone work in a pro studio and you can't get drunk or smoke during the time at all as it changes your voice however slightly and the result is ruined.

> Nowadays English vocal synthesis is a lot better and probably could do lead vocals on a song.

I don't think even the latest release Cyber Diva and Cyber Songman / Dex and Daina are there yet. The ability to add emotion is still missing and fixing that requires special talent which only a few Vocaloid producers, like MitchieM have. Somehow the monotonic singing style of Hatsune Miku is less disturbing or even cute in japanese, compared to english.