CHAPTER 3.3: TUNING BASICS
Tuning has become more prominent in the UTAU community over the past few years. It's what amkes or breaks a cover altogther. Users have become renowned in the community because of how unique their tuning style is.
What Exactly Is Tuning?
To preface, tuning in UTAU is pretty much an abstract concept. But think about how people sing their songs. What makes their singing unique, and how can you translate that in UTAU? This is what the tutorial is for.
For this tutorial, you will need the following:
- A UST (preferably an untuned one)
- A song for reference purposes (usually corresponds to the UST you are using)
- The pitch editor plugin for UTAU (link here
For demonstration purposes, I will be using the UST I made from the last tutorial.
The Anatomy Of Tuning
In UTAU, tuning usually comes in the form of pitchbends. The pitchbends in a UST vary in shapes or sizes depending on the part of the song. In many cases, the tuning replicates the original singing as close as possible or takes a diversion from it to make a unique style of singing.
Not all tuning styles are human-like or even realistic. Some are even robotic-sounding. There is NO right way to tune a song, but even a few pitchbends there and there will benefit the output of the UST you are working with.
The following is a set of pitchbends commonly used in UTAU tuning.
The most basic pitchbend in tuning. Usually gives the transitioning notes some depth. Doesn't really matter if you place it up or down.
Another basic pitchbend.
A VERY common type of pitchbend found in tuned USTs. Used to give the singing some personality. Comes in multiple variations.
A type of pitchbends that provides for more stylish singing methods. Again, this comes in multiple variations.
Vibratos also make up the tuning of a UST. They're variations in the pitch that look like sine waves, like this:
They are extremely useful for long, drawn-out notes, but also for short notes that don't really need pitchbends at all. I use vibratos a lot in my tuning.
If you right click on a vibrato and select Pitch..., you are presented this screen. This is where the vibrato settings are located.
It's also where you can add the vibrato. A lot of the parameters are self-explanatory, but here's a list of what the parameters do:
- Length: How long the vibrato should be (measured in percentages)
- Cycle: How many frequencies the vibrato should have (measured in ms)
- Depth: How large the vibrato should be (measured in... cents)
- In/Out: The fade-ins and fade-outs of the vibrato (measured in percentages)
- Phase: The offset of the vibrato (measured in percentages)
- Pitch: The pitch of the vibrato (measured in percentages)
Emulating Singing Styles
Tuning seems a bit difficult to learn, but using references HELPS A LOT. Trying to emulate one's singing style will certainly help you in developing your own tuning style. Hell, even looking at tuned USTs by other people will benefit you.
At this point, you should have the pitch editor plugin ready. UTAU has a dedicated plugins folder, so when you download a plugin, put it into the folder for it to show up in UTAU.
I'm bringing back my blank UST from the previous tutorial, now that it has lyrics stuffed into it. So what next? I also have the original song and vocals, which I'll be using as a reference to emulate the original singing.
I recommend tuning the UST in parts so that you don't get too lost where you left off. The image above is the region I'm working with. Select the region you want to tune, then go to Tools(T) > Plug-Ins(N) > 拡張ピッチエディタ. If that doesn't show up, click on Reload(R) and try again.
Now this screen should pop up. For some reason, the note lyrics are in mojibake if they're in Hiragana, but I think as long as you what the lyrics are, it shouldn't be that much of a problem. I don't even know if there's a fix for this yet.
Make sure that you have the [+] button toggled. That's the tool we'll be using to create the pitchbends according to the original singing style in the song you have referenced.
This part will take a lot of trial and error to make it sound right. Like I said, using references will HELP in you developing your own tuning style. Just use the four pitchbends I mentioned to make up the tuning or make your own pitchbends instead. Either way works, and there is no right or wrong way to tune a song. Add pitchbends depending on the singing you are referencing.
This is what my tuning looks like in the pitch editor...
...And this is what it looks like in UTAU.
Sure with have the pitchbends in place, but what about the vibratos? I said that vibratos are often used for longer notes, but they can be used in place of pitchbends at times or along with them. If the song I'm referencing has vibratos somewhere, then I will place vibratos within its respective places in the song.
Repeat the process for the remaining regions until the entire UST is tuned.
Here's what the UST looks like with pitchbends and vibratos:
How does the tuning sound, and how does it differ from an untuned UST? There is a major difference when you render a tuned UST and then an untuned UST. The following audio clips use Tokumei.
The untuned UST sounds kind of bland, while the tuned UST has a more emotional singing style and character to it. Other than that, congrats. You've tuned the main melody of the UST! You can repeat the process for the other USTs, but as for harmonies, I don't think you need to tune them unless you really want to.
The next part will cover on how to mix a cover.