Welcome Guest [Log In] [Register]




Welcome to the UTAforum!

UTAU is a Japanese voice-synthesis program used to create singing vocals for music. It is much like the more popular, commercially sold Vocaloid products, but UTAU is freeware!

To learn more about UTAU and various UTAU voices/characters, you can click here.

You're currently viewing our forum as a guest. This means you are limited to certain areas of the board and there are some features you can't use. If you join our community, you'll be able to access member-only sections, and use many member-only features such as customizing your profile, sending personal messages, and voting in polls. Registration is simple, fast, and completely free.


Join our community!


If you're already a member please log in to your account to access all of our features:

Username:   Password:
Add Reply
Smeen's Beginner Tutorial; That is way too big and image heavy
Topic Started: Nov 30 2011, 08:23 AM (748 Views)
smeen
Member Avatar
Momma Smeen
Welcome to this insanely big UTAU tutorial. This has been made for people who are either new to the program or are having problems with it. Advanced users, however, are free to add parts to this tutorial as they wish. To the people following this tutorial, follow every step completely and precisely and UTAU should function normally. This has quite some steps though, so be patient and keep your focus. If you still have problems or questions afterwards, be free to contact someone. Now, let’s commence, shall we?

Step one: Preparing your computer
NOTE: This tutorial is made for Windows users! If you are on a Mac or Linux I can’t help you. Mac or Linux users are free to post a different version of all of these steps, I will incorporate them in this tutorial if I can get them!
Preparing your PC is very important. You do this before you download UTAU. This is because UTAU might not work correctly if you download it before preparing your computer. So, what do you do?
Change your locale, or download AppLocale.
NOTE: As I do not use AppLocale, I’m going to rely on all the information I have read about the program. I do advice changing your locale, because AppLocale still might cause bugs and does not give you the ability to use voicebanks encoded in kana.

Changing your locale does not change a lot at your PC, however, some programs might run their installer in Japanese. I only had this once, however, so you should be considerably safe. These images show you how to change your locale on Windows 7. Windows Vista and XP should have similar steps on changing it.
Spoiler: click to toggle

Now, your PC will need to restart. Just let it restart.
Downloading AppLocale should just go as the downloading-page says. Install it through the installation program and you’re done. You can download it here: http://www.microsoft.com/download/en/details.aspx?id=13209

NOTE: European users need to do something else before running UTAU their PC is prepped correctly. People from the US, Australia and Asia can skip this step. I don’t know about people from other regions in the world. If you want to be sure, just check this step.
European users need to change another tiny thing in their system settings. This change has absolutely no consequences for your PC, but it has a big, big importance in running UTAU. These images show you what should be done.
Spoiler: click to toggle

That’s right, you change the decimal-sign to a dot (.) . This is absolutely necessary.

After you’ve done this, it’s time to get UTAU.
Edited by smeen, Nov 30 2011, 08:40 AM.
Disclaimer: Anything this smeen can say or do can be very, very strange.
Offline Profile Quote Post Goto Top
 
smeen
Member Avatar
Momma Smeen
Step two: Downloading UTAU (and a voicebank) and installing it.
This step is fairly easy. Downloading and installing UTAU is done before you know it. You download UTAU at this site http://utau2008.web.fc2.com/ . This image shows where to click.
Spoiler: click to toggle

If you click this page, you get on a page with a lot of links and Japanese text. Fear not, clicking the first link beneath Download will let you download the installer, and the second link will let you download the .zip . Here is another image to clarify.
Spoiler: click to toggle

Downloading the installer will run you through an installation program that automatically installs UTAU. Downloading the .zip will give you UTAU as a .zip file which can be extracted anywhere you please. UTAU does not have to be installed or extracted in your Program Files-folder, you can set it anywhere you like.
NOTE: AppLocale-users need to download and install UTAU in Japanese locale to get it to work!

Next thing to download is a voicebank. UTAU comes with Uta Utane (Defoko) as default bank, however this bank is poorly configured and hard to use (AppLocale-users will probably be unable to use her). This is why I advise to download a different bank. For beginners it’s probably the easy to download a voicebank that is encoded in romaji and aliased in kana. Applocale-users have this as their only option, seeing they cannot use banks encoded in kana. Most voicebanks come in .rar or .zip and need to be extracted. There are two ways of getting this bank into UTAU, and those start at extracting your bank. I will show you.
Method one:
NOTE: This method has been proven to give errors whenever UTAU has been installed or extracted into your Program Files-folder. Not everyone gets these errors, but to be safe I redirect you to method two.

As you see, your UTAU-folder has a VOICE-folder. You can extract voicebanks into this folder to get them into UTAU automatically.

Method two:
Make a folder for voicebanks where you want and extract the downloaded bank there. Opening it in UTAU will take some more work, as explained in these images.
Spoiler: click to toggle

This way you will have your voicebanks in the UTAU program just like method one.

NOTE: There are currently a lot of translations for UTAU. I will not explain how you can get UTAU in English or another language, to learn more about this process please go here: http://utau.wikia.com/wiki/UTAU_wiki:UTAU_GUI_Translation . I’ll be continueing this tutorial with an English version of UTAU.

Now we’ve got this covered, let’s look at recording your own UTAU.
Edited by smeen, Nov 30 2011, 08:26 AM.
Disclaimer: Anything this smeen can say or do can be very, very strange.
Offline Profile Quote Post Goto Top
 
smeen
Member Avatar
Momma Smeen
Step three: Recording a voicebank
NOTE: Please, please use a decent microphone when recording. It is not necessary to buy something expensive, but a built-in microphone is not good enough. This is not only to guarantee a good quality sound, but it will also make your voice more distinctive and your samples easier to oto. And to be honest, most people prefer good sounding samples over bad sounding ones. Just saying.
There are a lot of different programs to record in. Just to name a few, Oremo, Mixcraft, FL Studio, Audacity, Reaper and Garage Band. Whatever program you use, the quality will not differ very much, despite what some people say. I refer to scarfu’s comparision video for proof. I will not cover how to record with pictures, instead I’ll name some basic things you need to know.

NOTE: Since this tutorial is made for beginners, this will only cover creating a CV voicebank. I’m going to be mean right now and state that beginners should not touch VCV, CV/VC and CVC until they are familiar with CV. Because these methods are not made for beginners. And no, I’m not apologizing for this statement.

Make a folder for your samples before recording. When recording, there are things to think of. These are important to optimize the sound quality of your samples. Please read them through and check them off.

- Record your samples in Mono. One-track samples. I guess everyone knows what I mean by this. UTAU does not like stereo-samples and will most likely kill your UTAU if it has stereo samples.
- Save your samples in .wav format. It’s the best format for UTAU.
- Make your samples around 2 to 4 seconds long. This makes it easier to oto them afterwards.
- If recording Japanese and you can’t read hiragana, save your samples with romaji names. You can add hiragana aliases to the oto later.
- Your recordings all need to be done in the same pitch, tone, vocal impression.
- Vowels that pop, like b,p,t and ch, will sometimes make your sample pop too. You can solve this problem by buying or making a pop filter, but recording through a piece of cloth usually does the trick too.
- Get yourself a list of sounds for a basic voicebank in the language you want to record in. Japanese is the easiest since it hasn’t got a lot of samples. English, however, needs a hella lot of them. For the ease of it, this post will contain a list of sounds for Japanese voicebanks. If you are using Oremo it can be loaded into the program directly.
- This is pure for aesthetics, but dear beginner… Don’t use English r’s… Please try to learn the pronunciation of Japanese a little before recording.

NOTE: The reason why I advise people to use Oremo is because of the speed. Oremo can load a list of samples to record, set the destination folder and you’ll be done in a spiffy. However, since it’s a Japanese program it needs the same things as UTAU when using it in AppLocale. Also, as I mentioned earlier, no quality difference.

I already explained how to get a voicebank into UTAU, so I won’t go over this again. After you got it in UTAU, you can proceed with the next step.

NOTE: For those who want that list of sounds, here it is:
Spoiler: click to toggle
Edited by smeen, Nov 30 2011, 08:29 AM.
Disclaimer: Anything this smeen can say or do can be very, very strange.
Offline Profile Quote Post Goto Top
 
smeen
Member Avatar
Momma Smeen
Step four: Programming your recorded voicebank
After recording and getting your voicebank into the UTAU-program, you have to open the configuration-screen. I’ll show you how in these images.
Spoiler: click to toggle


If everything went right, you should now have a window like the last picture, but without any numbers or circlers behind the names of your files. Those circles represent the presence of a .frq file for your samples, the numbers are your oto.ini settings. Both are necessary to get your voicebank to function properly. We will render .frq files first, so you can take a break. Because…

NOTE: Rendering .frq files may take ages. Make sure you have enough time to let them finish and get yourself a cup of tea or something in the process. You can do something else in the meantime, opening a different program, even a game, will not interrupt the process.

These images show you how to make .frq files for your voicebank.
Spoiler: click to toggle


NOTE: If you already know how to change a resampler, please know that only the regular resampler (resampler.exe or .dll) creates .frq files. Frame resampler (fresamp) uses .frt and TIPS uses .pmk files. Make sure you have .frq files in your voicebank when you first make it. Don’t make the other files until your bank is completely done.

Just let the damn thing load. There’s nothing to speed up the process. All you can do is wait.
Disclaimer: Anything this smeen can say or do can be very, very strange.
Offline Profile Quote Post Goto Top
 
smeen
Member Avatar
Momma Smeen
And if you’re done, it’s time to make an oto.ini! Happy happy joy… Most people see this as a terrible and boring thing to do. I think it’s rather relaxing. To put it simply, making an oto.ini might make or break you. Good thing is though, this does not need to be finished in one go. Whenever you need a break, just click the OK-button in your configurations-screen (the one we opened earlier) to save your progress for next time. There are different ways to make an oto.ini and I will only explain one way, but it is a way that I have been using for quite a long time.

NOTE: Again, this tutorial is only for CV voicebanks. This oto.ini method cannot be used on any other kind of bank. For tutorials on oto’ing VCV, CV/VC or CVC please refer to other tutorials.
You need to open the window where you can configure a sample. I will show you how here:
Spoiler: click to toggle


First, you need to know the basics, and for this I will show you a sample that has been configured.
Spoiler: click to toggle

Now this is explained, it makes it really easy to explain how you should oto. I will explain it with text only, now you saw the editing-screen. Vowels and consonant-vowels are programmed differently.

Vowel (a,e,i,u,o,n)
- The red and green line are left alone. Their values are supposed to stay 0
- The blue at the beginning should cut off any excess recordings and stop when the actual sample starts.
- The blue at the end should cut off the ending of the note. Drag it to where your sample is getting weaker.
- The pink should cover the beginning of your sample until it becomes consistent in size.

Consonant-vowel
- The red line should be between the consonant and the vowel.
- The green line should be between the blue area at the start of the sample and the red line. For short consonants it should be in the middle, for longer ones around 1/3 down the way.
- The blue at the beginning should cut off any excess recordings and stop when the actual sample starts.
- The blue at the end should cut off the ending of the note. Drag it to where your sample is getting weaker.
- The pink should cover the beginning of your sample until it becomes consistent in size.

I understand this might be hard, so I made two small videos to show you what I mean.
Spoiler: click to toggle


NOTE: If you still find it hard, note that there are ways to make it easier. In your sample-configuration screen, the P-button will play the edited sample and the + and - buttons will zoom in and out.

NOTE: There are different ways to record and oto breaths. These are different for everyone, so I will not discuss them here in detail. You can also leave them un-programmed if you wish.
Disclaimer: Anything this smeen can say or do can be very, very strange.
Offline Profile Quote Post Goto Top
 
smeen
Member Avatar
Momma Smeen
After your oto.ini is completed, it is time to add aliases. Basically, it will make your UTAU able to sing in both kana and romaji. If you have a voicebank encoded in romaji the aliases will be in kana and if your voicebank is encoded in kana the aliases will be in romaji. Aliasing can be done by hand, but there is an auto-aliaser that makes everything much easier. I will show you how to add kana aliases by hand, though.
Spoiler: click to toggle


Be sure to check your aliases if you aren’t sure of what you’re doing. And if you’re done, it’s time to test your voicebank!
Disclaimer: Anything this smeen can say or do can be very, very strange.
Offline Profile Quote Post Goto Top
 
smeen
Member Avatar
Momma Smeen
Step five: Making a voicebank sing.
This is the last step in this tutorial, but it’s still important. Most people tend to forget a lot of things when opening a .UST or importing a .VSQ or .MIDI so I will show you the exact steps you should take to make your voicebank sound the best.

Opening a .UST is the easiest thing. These images will clearly show you the steps to take if you want to use a .UST .

NOTE: Of course you need to download a .UST before you can actually use it. They usually come in .rar or .zip format and can be extracted anywhere you like.

Spoiler: click to toggle


NOTE: Images continue in the next post due to image-limits!
Disclaimer: Anything this smeen can say or do can be very, very strange.
Offline Profile Quote Post Goto Top
 
smeen
Member Avatar
Momma Smeen
Spoiler: click to toggle


NOTE: Most .USTs will come in parts. You have to open each part individually.
Edited by smeen, Nov 30 2011, 08:40 AM.
Disclaimer: Anything this smeen can say or do can be very, very strange.
Offline Profile Quote Post Goto Top
 
smeen
Member Avatar
Momma Smeen
Usually, .VSQs and .MIDIs need some more work before you can use them. And I have a couple of notes to add to this before I’ll show you how to work them.
NOTE: .VSQs can be opened by UTAU in gibberish. There is a way to work around this, explained by Aster Selene here: http://overseasutau.forumotion.net/t2731-fixing-a-gibberish-vsq.
NOTE: Some .MIDIs might not work in UTAU. Others may need lyrics inserted.
However, all .VSQs and .MIDIs need one common thing. Decent stringing and pitchbends. And this is how you import a .VSQ or .MIDI and add those strings and pitchbends. After you have done that, repeat the thing you did with the .UST after you opened it. I’ll show you what I mean.
Spoiler: click to toggle


NOTE: Images once again continue in the next post!
Disclaimer: Anything this smeen can say or do can be very, very strange.
Offline Profile Quote Post Goto Top
 
smeen
Member Avatar
Momma Smeen
Spoiler: click to toggle


NOTE: One more post with images!
Disclaimer: Anything this smeen can say or do can be very, very strange.
Offline Profile Quote Post Goto Top
 
1 user reading this topic (1 Guest and 0 Anonymous)
ZetaBoards - Free Forum Hosting
Enjoy forums? Start your own community for free.
Learn More · Register Now
Go to Next Page
« Previous Topic · Tutorials and Resources Directory · Next Topic »
Add Reply

- Affiliates

- Chatbox

Welcome to the Cbox! Please be sure to follow the rules of the forum when posting. To use an avatar, simply upload a 45px by 45px image to an online host, and paste the direct link to it in the URL/Email blank next to your name!

• No flooding/spamming.
• No roleplaying in the Cbox.
• Your Cbox username must match your forum username.
(not exactly, but close enough that other people can tell who's who).