Resource icon

Tutorial Tuning Resources

This is going to be a collection of information regarding tuning, the process of moving from a melody with inputted lyrics to your desired finished sound. This can include things like adjusting the Vocaloid parameters (GEN, CLE, BRI, DYN, VEL, etc.), swapping out phonemes, etc.

Although I have some thoughts on the subject, I'm not an expert tuner, so most of this guide will likely focus on providing links to those who have provided more in-depth knowledge on particular areas. If anyone has additional resources to add, please message me or leave a comment! I'll be sure to give you credit for the items you provide, likely either by a quote or @ mentioning you in parentheses.

General Thoughts on the Tuning Process
When you first get into the Vocaloid interface, it looks like there are a lot of things to play with, and it's easy to feel overwhelmed at first. Below are a few thoughts on how to move on from there and approach tuning in general.

I think a good way to learn how to tune is to simply do tests with each parameter enough so that you are used to them and know what each does when turned up high and low.


I usually start blocking in what I want stuff to sound like the same way for every song. So I use DYN to make the end of notes taper out, or use it to increase the volume at the start of a note for more power. Then add some gender factor to help notes sound more masculine or feminine (a lot of people draw it as a straight line, but you can actually draw it in wiggley amounts because certain vowels sound weird with too much). Then do some flair stuff like add breathy sounds to the voice or Pitch Bend so you can decide how warbly they sing. Basically draw mountains and wiggles and keep experimenting.
I agree with everything uncreepy said. I think the right approach is to try to get a feel for what each parameter does--read the descriptions in the manual (which sometimes aren't clear right away, but they help), put in a test note and turn each parameter up and down on its own, try to get a feel for how each parameter changes the voice. Then, when you're creating, block in the vocal by putting in the notes and the lyrics so that your melody is basically there, and then go through the thing carefully and try to imagine how the voice is changing/what the parameters might be doing. You don't necessarily have to change every parameter all the time, just go for the big ones first--is the voice getting clearer on this note (parameter: CLE)? Brighter (parameter: BRI)? Is there kind of a distortion/gravellyness/growl to it (parameter: GWL)? Is it getting louder/quieter (parameter: DYN)? Try to capture the bigger changes first, and if it still doesn't sound how you want, try to notice the littler things (the parameters that aren't changing as much) and pick at those, probably by smaller amounts than the big changes.

It's a little esoteric and sentimental as advice, but don't think of your Vocaloid's voice as a complicated mess of things that you have to subdue. Think of it like sculpting pottery--create a general shape and then gently, patiently guide and smooth it into your desired form.

Guides to Tuning

Whatever Vocaloid editor you're using may have come with some kind of documentation to help users get the most out of the software. Both Piapro Studio and Vocaloid 5 definitely do. Piapro's is included with the software as a PDF. Vocaloid 5's is built in and can be found in its menu (@uncreepy).

YouTube Channels
  • ボーカロイド公式 調教完全テクニック(@uncreepy, @Narumo Luckycarrot (Wow, I just rediscovered this thread on VVN; I had completely forgotten that Narumo Luckycarrot had re-created it here from VocaloidOtaku. Here's a link to that thread, too. I'll leave this here, though, since the list of the chapters is useful.))
    • This book was originally written in Japanese by Pepperon-P, but Narumo has created English-translated versions of many of its sections. It is based on Vocaloid 3, but the principles generally still apply today. It covers a BUNCH of tuning topics, with both simpler principles and little notes on more advanced effects. I've provided a list of links to the translated sections below:
Part 1

Extra 1 (Guess this goes here?)

Part 2

Part 3

Part 4

Part 5

Part 6

Part 7

Part 8

Part 9

Part 10

Part 11

Part 12

Part 13

Part 14

Part 15

Vocaloid Parameters

Phoneme Replacement
While the ability to enter natural language words into the editor is a valuable, time-saving tool, sometimes you may need to adjust the phonemes your Vocaloid sings manually in order to get the pronunciation you want.

Lists of Phonemes/Notes on Various Languages
(Many of these are from Vocaloid Wiki, which has lots of in-depth language/tuning information. Reviewing it is recommended if you want a LOT more detail.)
Phonological Transformations
This article on amino discusses how to tweak your phonemes to get a more "everyday-speak" kind of sound (e.g., pronouncing "have to" like "hafta").
Cross-Synthesis (XSY)
This is a feature where two compatible voicebanks are combined to produce a new sound. It's available only to Vocaloid 3 and Vocaloid 4 generation voicebanks. This feature is not supported in the Vocaloid 5 editor. It does remain supported in the Vocaloid Editor for Cubase. For usage/technical details, see this article on Vocaloid Wiki.

Enhanced Voice Expression Control (E.V.E.C.)
This feature strengthens or softens the vowel (Voice Color feature) or initial consonant (Consonant Pronunciation/Extension feature) of a note. It can also be used to control the amount of breath at the end of a note (Voice Release feature). It is exclusive to Crypton vocaloids, and only Japanese voicebanks can currently use it. See this thread on VVN for a concise explanation, or this topic on Vocaloid Wiki for a very in-depth explanation with usage instructions.

Note-Bending/Portamento/Syllable-Splitting and Pitch Bending
These are two techniques for making a Vocaloid's voice "slide" from one note to another. In the former, you use dashes to connect the phonemes assigned to particular notes together. This lets you take a single syllable and stretch it over multiple notes. It can also be used to add little instances of higher/lower pitches before or after a note. In the latter technique, pitch bending, you use the Pitch Bend (PIT) and Pitch Bend Sensitivity (PBS) parameters together and literally draw the shape of the path you want the Vocaloid's voice to take in the editor. This is commonly thought to produce a more natural effect, but it's the more difficult of the two techniques.

@VegetalJuce made a great comment on Vocaloid Amino that explains this. You normally have to be logged in to Vocaloid Amino to be able to read comments, but I'm going to hope no one minds if I reproduce it here (reorganized a little to put the pictures in the right places). I believe it's based on the Vocaloid 4 editor, so there may be some minor differences from other editors here and there (@VegetalJuce).

Pitch bend! Yes, I use it all the time. This might get very long so prepare yourself lol:

First off there are 2 ways to "bend" the pitch. The most common one is called "note bending". This is when, for example, you might see people place a short note before or after a word, above or below it. The purpose of this is to make the voice start/end at a higher/lower pitch than the majority of the word. (SEE PIC #1)

You see this a lot in the form of a short higher note after a word, to do a cutesy upwards bend. To do this type of pitch bending, you have to use a dash (portamento) on a connecting note to carry the phoneme you originally added through all the rest of the notes.

Most people prefer this method because it is easy to see "where" the voice is going, because you are actually placing a short note at the desired pitch. It gives more "control".

Can sometimes negatively affect pronunciation, and it's easy to accidentally throw off the timing of the syllables when you're adding all these extra notes.

The second method is by using the PIT and PBS parameters. I'll go right out and say that this is my preferred method. This is a little harder to get the hang of, but I feel it lets you be more precise, creative, and detailed. To use this method, you first need to understand how the parameters work. PBS stands for "pitch bend sensitivity" and the default setting is 2. PBS controls how much the PIT parameter affects the voice. If PBS= 2, when you draw all the way to the top/bottom in the PIT, the voice will shift up/down two bars, to the next bar of the same color. If you set PBS to 4, drawing all the way to the top/bottom in PIT will make the voice go up 4 bars, or to the second bar of the same color, and so on. (SEE PIC #2)

@mobius017's note: I believe "bars" in this context are the black/white keys on the piano roll. In other words, tones and semi-tones. It's important to remember while drawing your pitch bends with the PIT parameter that the PIT parameter ranges from 8191 to -8192, with the unaltered pitch at 0. So when drawing, you can use those numbers as a guide for where the different tones/semi-tones fall. Just take 8191 or -8192 (depending on if you're bending to a higher or lower pitch) and divide it by the number of semi-tones you're going to be moving (The maximum number of semi-tones you can move, incidentally, is the same as the number to which you set PBS earlier.). The number you come up with will correspond to where one semi-tone is in the PIT parameter; each subsequent semi-tone higher/lower will be a multiple of that.
Allows you to have more variety and detail with how the voice bends around notes, helpful for slow & long transitions, will never mess up pronunciation or timing.

Takes some practice to get the hang of, requires a little more concentration to do. Involves 2 parameters, which can be hard at first.

Now let me give you some tips I learned the hard way:

1. DON'T USE THEM BOTH TOGETHER(on top of each other)!!! It screws things up because they effect each other. This isnt to say you can't use them both at different points in a song, because you definitely SHOULD! Both types are better at certain things and you should utilize both when needed, just not right on top of each other.

2. When using PIT, you need to keep in mind that the program automatically compensated for natural pitchbend between notes and you will need to counteract that at times. To do this, you often need to draw "triangles", not just parts that have one flat side, like THIS: (SEE PIC #3)

3. Use pitch render to visually see the bends. You can find this under the "part" menu.

Soooooo.... that's all I can think of. I'm no pro, I'm actually kind of a n00b, sorry if this isn't super helpful. I know I have a lot to learn myself.
First release
Last update
0.00 star(s) 0 ratings

More resources from mobius017

Latest updates

  1. Pitch Bending

    I've added an update to the Tuning Resource that gives info on pitch bending.