FullMetal*Panic!

Posted: **Thu Nov 22, 2007 9:04 pm**

http://www.youtube.com/watch?v=a5kDcIJ9i_A
http://youtube.com/watch?v=6-tL6eZ3FHI
http://youtube.com/watch?v=rxtyGbDgajM

Think Garageband for otakus. This Japanese software suite lets you plug in lyrics and melody and generates an "authentic-sounding" song via its music and vocal synthesizers. As you can see above, the software features a 16-year- old "Virtual Singer," which croons out whatever disgustingly sweet (or just disgusting) lyrics you enter in (Japanese only, we're assuming). It's so popular in Nippon that it's actually the #1 selling software on their Amazon. And for good reason—the songs they generate actually sound like it could have come from a generic teenaged anime. Hit the jump for two videos.

Yes you may hate me now ...

Posted: **Fri Nov 23, 2007 1:50 am**

oh, just what I need
as if the original loituma girl didn't enslave me at least once a day with its hypnotic leak spinning

now I have a techno version to listen to

but this is a pretty awesome concept anyway

Posted: **Fri Nov 23, 2007 2:31 am**

Yamaha has always been pouring tons of research into formant synthesis and this is the latest incarnation.

There are 3 variants (IIRC), Miku, Lola, and Leon. Lola is an older, english speaking singer.

lola: http://www.youtube.com/watch?v=SWLXoxF02Qg

Yamaha's site http://www.vocaloid.com/en/introduction.html

more samples
http://www.zero-g.co.uk/index.cfm?articleid=802

Posted: **Fri Nov 23, 2007 2:57 pm**

make a death metal voice and I'' buy it

Posted: **Mon Nov 26, 2007 1:39 pm**

Thats pretty crazy. You think they model the voices after existing people? They get royalties? Though not perfect, seems like it could help song writers... well, not much.

Posted: **Mon Nov 26, 2007 11:54 pm**

VOCALOID uses Frequency-Domain Singing Articulation Splicing and Shaping, a vocal (singing-voice) synthesizing system developed by Yamaha.

With this system, the "singing articulations" (collections of voice snippets, such as of syllables, and snippets of vocal expression variations like vibrato) needed to reproduce vocals are collected from custom-produced recordings of accomplished singers and put into a database after conversion into frequency domains.

To synthesize vocal parts, the system retrieves data consisting of voice snippets, applies pitch conversion, and splices and shapes them to form the words of a song as input by the user. As this processing is done at the frequency-domain level, pitch can be easily changed according to the specified melody, and the voice snippets can be spliced in a way that reproduces smooth-flowing words.

Posted: **Tue Nov 27, 2007 7:04 am**

This works for them pretty well because Japanese pronunciation is very simple usually one character has almost always just one pronunciation.

Posted: **Tue Nov 27, 2007 9:12 pm**

sounds pretty good, adding to my wish list.

FullMetal*Panic!

What in the ???? hell is this ... vocaloids....

What in the ???? hell is this ... vocaloids....