Vocaloid

[Steve]

I have just been reading the blurb for Yamahas new range of 'Vocal' samplers.

"VOCALOID is a vocal-synthesizing software that enables song writers to generate authentic-sounding singing on their PCs by simply typing in the lyrics and music notes of their compositions. The software synthesizes the sound from "vocal libraries" of recordings of actual singers, retaining the vocal qualities of the original singing voices to reproduce realistic vocals. The software also features simple commands that enable users to add expressive effects - such as vibrato and pitch bends - to their synthesized vocals. Additional releases to the range of vocal libraries currently available will broaden the range of voices and singing styles that can generated by VOCALOID. At the moment, VOCALOID can generate singing in Japanese and English. Other languages are also planned for the future.

VOCALOID software runs on Windows 2000/XP. A MacOS version is planned for the "

I have been hearing of this development for some time and was interested to see what it could do, SO... I went along to

http://www.timespace.com/index1.asp

To listen to some demos of this new software.

I downloaded the first demo, 'Check it out'. This is a very short piece of some 33 seconds. It's OK... Nothing to write home about..

So I downloaded the second demo 'Little bird'. I was absolutely blown away!!!! Surely this was impossible? The vocal was stunning!! I played this track through several times... I couldn't believe it? I must have more of this!

I went back for another demo, this time 'Lola is here'

What happened??? ... This is rubbish compared to Little Bird...Back for another download...

This one entitled 'Motherless Child.

Again ... Most dissapointing (compared to Little bird)

Back for another.. 'Introducing Lola'...??!??!

Then I found the track descriptions..... Little bird was the only track to have a real singer singing the lead vocal....

I was ready the sell the wife and kids into slavery for this piece of software!

Having said that! I think it could be usefull for harmonies and backing.

Check out the demos on the link obove and see what you think!

[Marc]

Are you kidding me? This is what I need!!! I was actually planning to make a soundfont using all the possible vowels and consonents to shape vocal articulations and use them in my songs. This might be better! However, I'll check the price before....

[Marc]

It sounds very processed. Too much like Cher in "do you beleive". Too mechanical. In some other song, lola sounds like someone who is mentally challenged and can't clearly pronounce some sounds.

I think the best sampled vocals are chorals, chanting simple latin words repetitively.

[BongStuff]

Quite interesting but I don't really want to sound like Daft Punk or Air in all my tracks.

Pretty scary what it could be in 5 years time though.

Add in some algorithmic composition, and you've got auto-generated pop-trash!

 BS

[Marc]

I'm impressed with these samples, that can be edited with a programs called "word building". Too expensive though,  $500 US.

http://www.soundsonline.com/cd_mp3_demos/141273.mp3

http://www.soundsonline.com/cd_mp3_demos/141274.mp3

[john]

Just to say, this post keeps getting picked up for people looking for a vocaloid crack.

Cheers

John

[Stefanos]

I had the same, i listened to that song and thought that's real. And it was  

Guess it will take years to get some reall vocals singing on the pc...........

[e-n-felix]

"Vocaloid V1.0 Serial Number Generator" is what people really need..

atleast for the "VOCALOID Software Update version 1.0.2"

keywords.. heh.. no sir.. i'm not looking for CRACKZ..

no WAREZ for me..

i can't even find the original binary..

#audiowarez on IRC seems to be locked up pretty tight..

what's the secret password?

*clicks his heels three times winks twcie and nudges the /join.*

and no mention of Vocaloid on the p2p nets..

I'm assuming it's released if they're offering updates..

meh.. if i did find one of the aforementioned scandelous items i'd surely hide em safely at:

http://www.e-n-felix.com/forum

so they wouldn't fall into the wrong hands ..

of course..

[e-n-felix]

Hit Song Science + Vocaloid = $.$

[e-n-felix]

   HSS statistically analyzises Top 30 hits to determine mathematical similarities in pitch, tempo, and structural format, as well as spectral analysis and over 25 additional variables directly from the digitalized audio signal.

   It has been noted that there are a finite number of clusters that all most all hit songs fall into.

   Although I have not had direct access to this data, it can be hypothesising that there are 18-55 mathematical archtypes based on the clustered formation.

   HSS does not take into consideration of vocal content although the tonal aspects make a marked expression as a vocal insturment. While Vocaloid may still be in it's infancy, I believe a synthetic vocal insturment will be more efficent at matching the model of the mathematical archtype.

   This voiceset sampling technique  is similiar to MBROLA synthesis. Voice set creation is extensive and requires a studio enviroment. the possiblity of reanimating a deceased performance artist vocals is highly unlikely with this technique. Leave it to Yamaha to dress it up fun enough for kids to play with.

Check out this topic bending HSS link:

http://www.hitsongscience.com/index.php?p=w

[Marc]

Hit Song Science + Vocaloid = $.$

Guess you're not interested in art and its place in music, then...  :'(

[e-n-felix]

I've spent close to two decades making music strictly for the sake of art and my own enjoyment. I've also been paid to play and dj before. But one must admit the reality of this is little more than a hobby.

   During my music theory training the rigidness with which the classical masters held to musical formula was really emphasized. From traditional chord progressions to song structure there was an underlying set of principles that steered these composers to produce masterpieces with a pronounced clarity of vision and appeal to the masses.

   I believe the statistical analysis of music aesthetics will yield way to another set of musical formula. The evolution in the theory of musical archetypes is still young but with the proper evaluation of all relevant data one should be able to distinguish the optimum structural and tonal aspects of music with mass appeal.

   As of yet these "Arch types" have not been created, no ones ever heard them.. we barely know they exist by statistical analysis. Almost every bodies favorite Top 30 song is mathematically similar to one of these archetypes, so in effect they would have a haunting familiarity.

   Arguably they could be considered the true core of modern musical art.

   And to hold to topic I think Vocaloid Technology could play a key part in engineering these beasties.

[Steve]

   HSS statistically analyzises Top 30 hits to determine mathematical similarities in pitch, tempo, and structural format, as well as spectral analysis and over 25 additional variables directly from the digitalized audio signal.

   It has been noted that there are a finite number of clusters that all most all hit songs fall into.

   Although I have not had direct access to this data, it can be hypothesising that there are 18-55 mathematical archtypes based on the clustered formation.

   HSS does not take into consideration of vocal content although the tonal aspects make a marked expression as a vocal insturment. While Vocaloid may still be in it's infancy, I believe a synthetic vocal insturment will be more efficent at matching the model of the mathematical archtype.

   This voiceset sampling technique  is similiar to MBROLA synthesis. Voice set creation is extensive and requires a studio enviroment. the possiblity of reanimating a deceased performance artist vocals is highly unlikely with this technique. Leave it to Yamaha to dress it up fun enough for kids to play with.

Check out this topic bending HSS link:

http://www.hitsongscience.com/index.php?p=w

So all we need is the formula and we can have a hit record in a blink of an eye!... We need to send in the Finnish comandoes to aquire the formula... H, Ollgu! Get yer gear on, you're going in...  

[Marc]

So all we need is the formula and we can have a hit record in a blink of an eye!... We need to send in the Finnish comandoes to aquire the formula... H, Ollgu! Get yer gear on, you're going in...  

Finnish commandos, MBROLA soldiers of fortune  and "Arch types" special units. Who's withholding the info? I bet it's the devil who also possesses Caramilk's secret... or it might be Kentucky's Colonel Sanders who, after sending soldiers to the civil war to be killed is now sending fried chicken with his secret recipe to bring people to a slow death!