For your very first voicebank, I highly recommend starting with CV Japanese. This will allow you to better understand the basic workflow and concepts of creating a voicebank for UTAU.
1. Preparation
You will need a recording program, a microphone, and a recording list.
Audacity is a good general-purpose program for basic audio recording and editing. OREMO and Akorin are both specialized programs for recording UTAU voicebanks specifically.
If you already have a microphone, you can use your current setup. If you don't, but you do have a phone, you can use it as a microphone with your computer using WO Mic:
http://www.wolicheng.com/womic/ . If you don't have a phone, but your computer does have a built-in microphone, that's fine for a first voicebank. We're learning the workflow, not trying to make the world's best voicebank.
Both OREMO and Akorin come with a CV Japanese recording list, or reclist. If you're not using one of those to record and need to find a reclist elsewhere, here's a guide on what makes a good Japanese reclist.
https://utaforum.net/resources/what-to-look-for-in-a-japanese-reclist-and-what-to-avoid.170/
Create a new folder to store your recordings.
2. Recording
Make sure you're in a quiet, isolated environment. Turn off sources of noise like fans or air conditioners. Let others in the household know that you need the quiet in order to record properly, and keep pets out of your room. If you live on a busy street, you'll have to strategically record samples in between noises from outside.
For a single pitch voicebank, pick a musical pitch in the middle of your range, or maybe a bit lower than the middle. (UTAU generally does a better job shifting a note upwards than shifting it downwards.) OREMO has a built-in pitch reference tool, but other than that, you could open UTAU and put in a single note to hear Defoko sing your reference pitch.
For each CV recording, you'll want to sing and hold the syllable for a few seconds.
If you're using Audacity, make sure there's only one track when you record, not two. The default sample rate should be 44100Hz. Make sure to export the audio as a 16 bit WAV file. OREMO and Akorin automatically handle the correct audio file format.
3. Configuration
Once you've finished recording, you'll need to configure the OTO so that UTAU can correctly time the samples and connect them together. Here's a guide on how to do that with UTAU's built-in OTO editor, for a CV Japanese voicebank:
https://utaforum.net/resources/cv-otoing-guide.269/
If you'd like to add a picture to the voicebank that'll show up in the top left corner of UTAU, it needs to be a 100x100 BMP file. Then you should create a plain text file named "character.txt" with the following contents:
INI:
name=Voicebank name
version=Version number
author=Your name
web=Your website
image=name_of_image.bmp
sample=name_of_sample.wav
If you intend on distributing this voicebank to the public, you should write clear terms of use for it. Here's a generator that'll help you with that:
https://tools.tubs.wtf/vbtougen
Once you've downloaded the readme.txt file, you can put it in the voicebank folder, and it'll be viewable inside UTAU when you click on the profile picture.
