Ah, speech synthesis... it can give voices to those with throat cancer, read webpages for the visually-impaired, and put established singers out of work. Wait... what?
Enter Vocaloid, a set of lyric synthesisers by Yamaha Corporation. Using the same general principle behind normal speech synthesis, Vocaloid can passibly duplicate any lyrics in any key range, using an impossibly large set of phonemes and syllables. In fact, as its primary usage is in the genre of techno, where every instrument is synthesised, it can become difficult to tell which singers are acoustic and which are Vocaloid. At some points, you can easily be fooled into thinking that a Vocaloid is simply a human using AutoTune.
However, trained ears can distinctly tell acoustics from synthesisers... Vocaloid cannot produce any vocal effects, such as hoarseness, whispering, and shouting during playback. It also cannot "speak" (that is, to deliver speech without music). Nevertheless, the programme remains very popular with techno composers in Japan and Europe (and the United States to a certain extent). YouTube is replete with anime Vocaloid techno and the programme's "mascot", Hatsune Miku, is just as recognisable in Japan as the Mario Brothers.
However, what does this prolific synthesiser say about the people who use it? Where is the talent in composing lyrics that you want a synthesiser to sing? Of course, consider that, in techno (a genre about which I, admittedly, know very little) the objective is to write music that is fully synthesised. Before the popularity of Vocaloid, one way this was done was simply to digitally change the pitch of a human voice or to use a VOCODER. However, the latter provided very little variety beyond the standard "singing robot" and, with the former, it is possible to pitch-shift too much, causing entire syllables to be lost in translation, curtailing the composer's freedom in writing high or low notes. Vocaloid, being based on human speech, provides the composer complete freedom to write whatever he desires -- it will be sung properly (though a certain degree of shifting is occasionally required). Also, it isn't a VOCODER, meaning the composer can abandon the "singing robot" archetype.
Beyond techno, however, I don't see what future Vocaloid can possibly have. Its lack of vocal effects or even dynamics (volume-changing ability) makes it unsuitable for opera or musicals... its Japanese-language basis makes it unsuitable for most varieties of hip-hop... its melodic sound makes it unsuitable for rock. The only hope Vocaloid has is if '90s pop is ever revived, with its somewhat haphazard melange of '80s synthpop, rock, and R&B. The Spice Girls and Britney Spears would have been prime targets for replacement, had Vocaloid been around then.
Still, it's good I'm a composer, not a singer. This programme's existence foreshadows the obsolescence of the live performer.
