Spirit World Part 2

I think i got the descend into Spirit World covered but i'm not sure if the next part should be with any beats but i have come up with something as a base for diving into part 2. My basic idea for part 2 is something melodic and perhaps minimal beat like but also with unusual sounds here and there. As it is now the drums which pops up seems natural to me and not disturbing the mood. The trick now will be to keep the mood and add more rhythmic stuff without creating stress.

http://www.michaelkastrup.com/synthdemos/MK_SpiritWorld_Part2.mp3

/Michael

Spirit World Part 1

Today i felt like making something very different, my initial plan is to make a 15-20 mins piece of music which is a blend of sound fx and chill/relax. I want it to be something that place the listener in another (world) so lets call it spirit world. It starts of a bit spookey but that is just to sharpen your attention. This first part i would like to think of as a descend into that world. Once there new stuff will happen (part 2 i suppose).

http://www.michaelkastrup.com/synthdemos/MK_SpiritWorld_Part1.mp3

/Michael

Inspirational Mega Patches

Last night i tried build a mega patch with the following ingredients:

Zebra = Low end string, Violins, Flute
Sampler = Cellos ensemble
Sampler = 7 Sweet Ann vowel samples controlled by crossfade velocity.

The harder i played the brighter vowels and more strings and flute would come in.

It turned out to be something i jammed with all night and couldnt stop, so today i decided to post one of the examples i came up with last night:

http://www.michaelkastrup.com/synthdemos/v2_sweetann_05.mp3

/Michael

Fourth encounter with Sweet Ann

After some hours i managed to put together some samples of various vowels. Now we are begining to see the possibility for choir stuff and i left out the initial backing choir which was not Sweet Ann. So this time its all Sweet Ann but sampled versions of her which is kinda cheat but its the only way to get the job done in my eyes. The drawback now is that voices come out like samples but i'd rather go with that than take the fight using Vocaloid2 and Cubase in Rewire mode.

Damn b#3?+ cant say You or Who, she says Yu and Hy and happely forgets about the bleeding O.

http://www.michaelkastrup.com/synthdemos/v2_sweetann_04.mp3

EDIT: By watching a tutorial i could see a method to get the "You" done, seems like i only need a very tiny part of the actual You word placed 2 semitones below a longer U word so the first You word kinda bends up to form a U.

/Michael

Third encounter with Sweet Ann

Ok, the pitch thing is due to several things in the way she is ment to sing and this makes it VERY difficult. The main problem is i cant compensate for the pitch when i have nothing to hold it up against like a rewire track (we tried that and it didnt work). So what happens is i find a phrase i like but when i have exported the singing to a wave file and want to try it out in my host it turns out that some part of the phrase is out of pitch, yikes. So now i'm trying to see if i can create a choir with various vowels by sampling each damn note and then fine correct for pitch on each note.

Then a new problem came up, it turns out that what you hear when you audition a phrase it might not come out the exact same way when rendered it to a wav file. For some odd reason Vocaloid2 decides to pick another of the singers basic samplings as a starting point for a word. A typical setup for a vowel is that its made up of 2 samples of the original singer. So you have a bright sample for the natural registers and another sample for the low registers.

In the following music piece Sweet Ann do the ae--ee--uu and another choir sample set do the intro choir. Infact Sweet Ann was supposed to sing ae--ee--uu--ahh--oh but the ahh and the oh came out dead wrong in pitch so i only kept the first parts.

http://www.michaelkastrup.com/synthdemos/v2_sweetann_03.mp3

/Michael

Second encounter with Sweet Ann

Ok i managed to find a pitch which had an overall good feel by using pitch bend in Vocaloid2. There is no overall pitch control in Vocaloid2 so using pitch bend was my only choice.

Now that the pitch problem is (solved) i can lessen the harmonizer effect to bring out the voice more clearly. I tried to put in some nonsense word just to get a feel of other things coming from her voice. I think the real hard work is making her sign the words in a correct manner and avoiding the weird articulations which is bound to happen on some words.

Here is a snippet of step 2 with Sweet Ann:

http://www.michaelkastrup.com/synthdemos/v2_sweetann_02.mp3

/Michael

First encounter with Vocaloid2 Sweet Ann

A while ago i picked up Vocaloid2 Sweet Ann and yesterday i decided to give it a spin.

One cool features is supposed to be the Vocaloid2 Realtime module which is quite handy but what ever you do with the realtime module cannot be rendered to audio. What a bummer, i thought i was on to something and wanted a snippet of the mix in Cubase with the Vocaloid2 Realtime thing. It turns out that the realtime module dosnt output anything when rendering even if you use realtime rendering.

Ohh well perhaps the rewire mode can be used to something then, and oops what a bummer. Having Cubase and Vocaloid2 setup for rewire mode made it impossible to record anything following a beat. I looks like the Vocaloid part is trying to compensate for voice intro and other stuff i dont know about yet.

Ok, how about i just used vocaloid2 and put in some hum and ohhs and THEN rendered that to a wav file. Ooops what a bummer, it turns out Sweet Ann comes out flat and out of pitch. Only way to get something useful of that was to have the rendered wav track run thru a harmonizer with lots of detuned voices as backup.

So is Vocaloid2 Sweet Ann worth it ? Are you kidding lol nopes unless you have way to much money on your hands. I do hope they will correct these bugs or what ever it is because as it is right now i could have more fun sampling a speaking doll.

IF i get something useful out of Sweet Ann i'll post it but that chance is pretty slim atm.

Ohh well, in all fairness here is a snippet where i'm making heavy use of harmonizer and delay effect because of the out of pitch voice:

http://www.michaelkastrup.com/synthdemos/v2_sweetann_01.mp3

/Michael