The Musicians' Corner
View Single Post
July 26th, 2012 (02:25 PM).
Join Date: May 2012
Location: In Kageroza's Arms
I work with a Korean rap/hip-hop artist now that I sing songs for, but other than that no. My recording programs have been giving me so much troubles. T_T
You can hear a small WIP sample of our first song here:
Playing with chorus harmonies and main melodies. It is a rap/rnb Kpop style. Title will be "Come Alive".
UTAU songs take longer/harder to make than me singing. XD;; But it depends on the quality of the available UST (that includes the notes data that the UTAU voice will sing) and whoever bank I work with on the song. Every UST is usually shipped ready to go for most UTAU voices, but Toad has a limited bank, meaning for many of his syllables I have to heavy edit the existing UST or make from complete scratch (as in that english one). He has roughly half of what a normal sized bank has.
An example from my Toad edits is that he doesn't have a "Ha" syllable, mostly because of his superly extreme accent which makes him saying "Ha" very hard to hear the H, so I had him record himself going "hhhh" and tack that on to the "a2" file, with "to" for the T. Here's a sample file of Toad singing the smoothest part of the song that I could get. I plan to rework much of that first portion to make it smoother. Mind you this is CV, not VCV so it will sound less realistic than VCV UTAU. But it's DANG close, I'll tell you that, because I'm slaving over pitch bends and envelops (the most fun part).
The real syllables you're hearing here are. "a2, n, du, su, be (he does not have a pe sound, it corrupts), se, su, bi, tu, wi, n, a2, su, yu, ha, vu, ka, mu, tu, sho, yu, go, a2, n" every syllable applied to the timeline and then edited with flags (which changes everything from gravel to breathiness to clarity), envelops, attack, volume, vibrato, pitch bending, etc. to make it smooth transition. You also must oto the file to fit some UST which tells the program how to render the note, where the consonant and vowel should start, and which part of the file to hold out/stretch. This becomes a problem for Toad's bank, which is limited so sometimes I will edit the UST piece by piece with different oto settings in order to give him clearer consonants or a more definite vowel sound.
I've put in around 7-ish hours on that from scratch UST. To modify a pre-existing one, I usually put in between 1-2 hours. That's including mixing.
"o.o What's that?"
Tha' means yor
. </3 T_T
View Public Profile
Send a private message to Kurui
Visit Kurui's homepage!
Find all posts by Kurui
Find threads started by Kurui
Ignore Posts by Kurui