Can someone explain this a little bit...?

Can someone explain this a little bit? Because when I read this, I only get a picture that I should make a note with three different octave..

e.g:

'a' --> 'a' in lower octave
--> 'a' in middle octave
--> 'a' in higher octave..

Am I wrong??

I'm really not certain about this but as no one is answering... ^^;;;

I do know there are Japanese UTAU voicebanks that record a high version of each phoneme... because I've had to take the extra character out of USTs to make parts work. So this makes sense with that...

That's as far as my knowledge stretches on this issue though.

Edit: Typos

dreamyilia wrote:I'm really not certain about this but as no one is answering... ^^;;;

I do know there are japanese UTAU voicebanks that record a high version of each phoneme... because I've had to take the extra character out of USTs to make parts work. So this makes sense with that...

That's as far as my knowledge stratches on this issue though.

*Sob sob*... Even though this silly question of mine had sunk for almost a week, you have pulled it up. Thank you for your concern. It makes me happy. >w<..

By the way, I'm curious about the definition of USTs and VSQs..

Cepom wrote:

dreamyilia wrote:I'm really not certain about this but as no one is answering... ^^;;;

I do know there are japanese UTAU voicebanks that record a high version of each phoneme... because I've had to take the extra character out of USTs to make parts work. So this makes sense with that...

That's as far as my knowledge stratches on this issue though.

*Sob sob*... Even though this silly question of mine had sunk for almost a week, you have pulled it up. Thank you for your concern. It makes me happy. >w<..

By the way, I'm curious about the definition of USTs and VSQs..

First things first, is the notes on different octaves thing.
You do NOT need to record separate phonemes in different pitches. This is only optional, and is REALLY REALLY REALLY optional for triphonic banks. This does, however, make each note sound relatively closer to your own voice, and can be used for a more high-quality bank.
I've not tested this with triphones before, but when Teto made her triphones, she didn't have other recorded pitches anymore, so I don't know the effect of other-pitched recordings on triphones.

Next. USTs and VSQs.
VSQs are Vocaloid SeQuences. They used in VOCALOID2 as the basic file type, however, VOCALOID MIDI can also be used, as these store the phoneme data and tempo.
VOCALOID MIDI is mainly used in VOCALOID1 and is usually ignored by those who use VOCALOID 2, however, it is handy to have one in handy in case you want to use VOCALOID 1 for some reason.

USTs are basically the same as VSQs, except they are made for UTAU. I'm pretty sure that UST is an acronym for "UTAU Sequence Text". USTs and VSQs are different from each other in two ways.
1. USTs can only have one track.
2. USTs have two "Modes"- Mode 1 (which is rather difficult to use) and Mode 2 (which commonly for japanese made usts, is not used). These modes can be switched by pressing the "Mode2" button in UTAU. Triphonic banks usually do not support mode 1, as pitch bends are made according to where each note's envelope begins, rather than where the note ends on the track.

*large wall of text is large*

Quote from LupinAKAFlashTH2:

*large wall of text is large*

You mean the image on my first post? My apologize, I don't know how to re size it.

Ahahaha. No actually my response to you. XD;

Hmm... that tutorial is really confusingly worded. o.o;; I think what the last part is saying is that you only want to record sounds that end in a vowel. But really, there aren't any Japanese sounds that end in a consonant, so if you're making a Japanese voicebank, you're not even going to run into that problem.

I'm sorry it took so long to answer your question~ T__T

LupinAKAFlashTH2 wrote:
Next. USTs and VSQs.
VSQs are Vocaloid SeQuences. They used in VOCALOID2 as the basic file type, however, VOCALOID MIDI can also be used, as these store the phoneme data and tempo.
VOCALOID MIDI is mainly used in VOCALOID1 and is usually ignored by those who use VOCALOID 2, however, it is handy to have one in handy in case you want to use VOCALOID 1 for some reason.

However, a much faster method is taking the VSQ from Vocaloid2 and dropping it over Vocaloid1's shortcut icon.

I cheat using that method for Vocaloid1 and KAITO.

Vocaloid1 can read Vocalod2 VSQs but as far as I'm concerned, dropping the VSQs on the Vocaloid1 icon is the only way they'll read it.

xLei wrote:

LupinAKAFlashTH2 wrote:
Next. USTs and VSQs.
VSQs are Vocaloid SeQuences. They used in VOCALOID2 as the basic file type, however, VOCALOID MIDI can also be used, as these store the phoneme data and tempo.
VOCALOID MIDI is mainly used in VOCALOID1 and is usually ignored by those who use VOCALOID 2, however, it is handy to have one in handy in case you want to use VOCALOID 1 for some reason.

However, a much faster method is taking the VSQ from Vocaloid2 and dropping it over Vocaloid1's shortcut icon.

I cheat using that method for Vocaloid1 and KAITO.

Vocaloid1 can read Vocalod2 VSQs but as far as I'm concerned, dropping the VSQs on the Vocaloid1 icon is the only way they'll read it.

Or you could just change the vsq file to a midi file.

I find VSQs a lot easier to use than Midi so I just use that method xD

Chasing Fireflies wrote:Hmm... that tutorial is really confusingly worded. o.o;; I think what the last part is saying is that you only want to record sounds that end in a vowel. But really, there aren't any Japanese sounds that end in a consonant, so if you're making a Japanese voicebank, you're not even going to run into that problem.

I'm sorry it took so long to answer your question~ T__T

Nah~ I don't really mind (or was I? :p). Does vowel mean like "ne" or somethinng like that? Or, or, or only the letter 'e' from "ne".

xLei wrote:I find VSQs a lot easier to use than Midi so I just use that method xD

I remember exporting VSQs from Vocaloid2 to UTAU.. is only opened the VSQs itself. Or exporting lyrics form UTAU by Midi. I forgot.

Cepom wrote:

Chasing Fireflies wrote:Hmm... that tutorial is really confusingly worded. o.o;; I think what the last part is saying is that you only want to record sounds that end in a vowel. But really, there aren't any Japanese sounds that end in a consonant, so if you're making a Japanese voicebank, you're not even going to run into that problem.

I'm sorry it took so long to answer your question~ T__T

Nah~ I don't really mind (or was I? :p). Does vowel mean like "ne" or somethinng like that? Or, or, or only the letter 'e' from "ne".

That's right. ^__^ In "ne", 'n' would be the consonant and 'e' would be the vowel.