Seeking some advice!

Hey guys,

My recording microphone's completely dead, so in the mean time I thought I'd ask a few questions regarding UTAU. I might as well start preparing for when I do get to recording.

See, I want to start out with a bang. I don't want to release the first act until I know it's well balanced. So my question is, how does one create a well balanced voice bank? From what I've noticed, a lot of the Japanese UTAUloids such as Teto and Momo have very charming and well balanced voice banks even for their first acts.

My first guess was to try referencing voice banks, but I was told that was a big no-no. I'm a bit confused as to why, but I'll take that person's word for it.

By no means do I wish to come off as a perfectionist, but I'd appreciate really good advice on making a solid voice bank that's not only pleasant to listen to, but pleasant to work with.

Thanks in advance!

What I do is to record everything at once and then separate them later. xD it really helps

KanashiiNoNeko wrote:What I do is to record everything at once and then separate them later. xD it really helps

I tried that a few months ago and it didn't end up so well. Thanks for the advice, though.

Zeny wrote:

KanashiiNoNeko wrote:What I do is to record everything at once and then separate them later. xD it really helps

I tried that a few months ago and it didn't end up so well. Thanks for the advice, though.

Huh? what happened? is it because Audacity crashed or something like that? Well yeah it's pretty boring to separate all the sounds after but...xD

KanashiiNoNeko wrote:

Zeny wrote:

KanashiiNoNeko wrote:What I do is to record everything at once and then separate them later. xD it really helps

I tried that a few months ago and it didn't end up so well. Thanks for the advice, though.

Huh? what happened? is it because Audacity crashed or something like that? Well yeah it's pretty boring to separate all the sounds after but...xD

That and when I redid it crashed again after trying to reduce the noise.

Zeny wrote:

KanashiiNoNeko wrote:

Zeny wrote:

KanashiiNoNeko wrote:What I do is to record everything at once and then separate them later. xD it really helps

I tried that a few months ago and it didn't end up so well. Thanks for the advice, though.

Huh? what happened? is it because Audacity crashed or something like that? Well yeah it's pretty boring to separate all the sounds after but...xD

That and when I redid it crashed again after trying to reduce the noise.

I can do it for you if you want, just send me it...I dunno...My audacity never crashed before

KanashiiNoNeko wrote:

Zeny wrote:

KanashiiNoNeko wrote:

Zeny wrote:

KanashiiNoNeko wrote:What I do is to record everything at once and then separate them later. xD it really helps

I tried that a few months ago and it didn't end up so well. Thanks for the advice, though.

Huh? what happened? is it because Audacity crashed or something like that? Well yeah it's pretty boring to separate all the sounds after but...xD

That and when I redid it crashed again after trying to reduce the noise.

I can do it for you if you want, just send me it...I dunno...My audacity never crashed before

No thanks, I'd like to do the recording myself when I get a microphone replacement. I might need help for the oto.ini file when I get to that point, though.

Canihelpplease?

KanashiiNoNeko wrote:Canihelpplease?

Sure, no problem, I'll look for help when I get to that point.

What Yanyu did with her vowel sounds (although this might not be new) to make it sound more natural is use the vowel sounds from Ka, Ki ect and just save them as i and a ect. (sorry if I don't make sense) it really makes a different then recording them on their own~<3 (Haha if you already know this the I have no more advice cause I fail at most stuff~orz)

Yanyu wrote:What Yanyu did with her vowel sounds (although this might not be new) to make it sound more natural is use the vowel sounds from Ka, Ki ect and just save them as i and a ect. (sorry if I don't make sense) it really makes a different then recording them on their own~<3 (Haha if you already know this the I have no more advice cause I fail at most stuff~orz)

Could you clarify that, please? I didn't quite understand.

So sorry~ Er... Well the A from the KA. Take that out so it just sounds like A make it a bit louder if needed. And do that for all the vowels~ Like Ka Ki Ku Ko Ke so you'll have all the vowels. It seems more natural when I did it because it flows more. Er..~ Did you get me this time? I'm sorry I fail so mecha...

Yanyu wrote:So sorry~ Er... Well the A from the KA. Take that out so it just sounds like A make it a bit louder if needed. And do that for all the vowels~ Like Ka Ki Ku Ko Ke so you'll have all the vowels. It seems more natural when I did it because it flows more. Er..~ Did you get me this time? I'm sorry I fail so mecha...

Ah, I get it now, thanks.

The best thing to do is just use OREMO.
And practice keeping your voice monotone the whole time.
The MOST IMPORTANT thing is to keep the same tone throughout. Second most important is keeping the pitch level within samples. Third most important is not varying volume too much.

annamaeblythe wrote:The best thing to do is just use OREMO.
And practice keeping your voice monotone the whole time.
The MOST IMPORTANT thing is to keep the same tone throughout. Second most important is keeping the pitch level within samples. Third most important is not varying volume too much.

I agree almost fully with Mae, although I am not sure keeping monotone helps. I am going to test this for a school project (hopefully), so I'll address that later, but the tone, pitch and volume are extremely important. ^^
Change any of these throughout recording and most likely your UTAU will be choppy.

Also, recording with a strong, clear voice always helps as long as your character isn't whispery.

Audacity works just as well for recording, though. Luna was recorded in Audacity, and people have told me she's smoother than Len, which I've compared them side by side. (foo, not trying to say she's better than everyone, that's just what i've been told, and I think she sucks. 8'D)
Namine Ritsu was also recorded in Audacity, so it's not really the recording program, it's more mic quality and your familiarity with the program you're using.
Like Yanyu said, you can use the same syllables, or for certain sounds that the vowels don't mesh well with, you can have special vowel sounds that are just the end of the syllable cut off to make it flow better. You just duplicate the syllable in your oto, but off the consonant, and alias it to be a vowel sound.
It's real easy~

Ah, alright everyone, thanks! I just bought a new microphone, hopefully I'll finish recording this weekend if I'm not too busy.

Check out my set up:

https://2img.net/h/oi50.tinypic.com/2vs5l3o.jpg

It's a photo of my old mic, but I have a new one of that exact model, so it's no different.

Ooh, nice! You have a pop screen, too, that really helps. ^__^ It helps if you play the pitch you're going to sing on a piano, then practice keeping your voice even on that pitch. And unless you're going for more of a soft voice, make sure to sing out. Good luck recording!