General

My configurations are Pay What You Want with a low minimum price. This is so that people who usually cannot afford to commission someone can do so at a low price, while people that have a higher disposable income are able to pay more so I can, you know, pay rent.

UTAU

I charge 50% extra for UTAU configurations that lack a base oto.ini. I do this because if a base oto is not present, it costs me a lot of time to seperate and assign the recorded samples to their UTAU counterparts.

To know your amount of recordings, go into the folder with all your UTAU's recordings. Select all audio files. If you're on Windows, it should show you ... items selected in the bottom left corner. This is the number you would fill in.

If you recorded with a reclist the amount of mora per recording is easy. If a reclist is advertised as 5 mora, fill in 5.
If you recorded with a reclist that has an uneven amount of mora, please calculate the average amount of mora per recording. I'm sorry.

Do you want to commission me for a configuration type or language not mentioned here? Please skip the calculator and go straight ahead to contacting me. I will calculate the minimum price for you based on your recorded voicebank. Once I've handled your commission I will add it to the calculator!

EN VCCV

I currently have a set minimum price for EN VCCV, assuming it was recorded with Cz's reclist. This price is per pitch. I know this is a lot, but EN VCCV is very very big so it's a lot of work :( If you're on a budget I would reccomend recording ARPAsing instead.

Diffsinger

I charge a little more for Diffsinger as I find labeling much more boring than otoing. Think of it as an anguish tax.

To know your dataset length, go into the folder with all your Diffsinger recordings. Select all audio files. If you're on Windows, right click one of the selected files and click "Properties". Go to the Details tab. It should show you the Length under Media. This is the number you would fill in.

If you want to avoid having an unreasonably high minimum price, please cut silences out of your dataset. This will make the dataset shorter and the price lower! ;)

I highly recommend including a transcription of your recordings. This will in most cases be song lyrics. If you do not provide this, I will have to use Whisper which provides less accurate results than a transcription would.

If you find a bug in one of these calculators, let me know please!

UTAU

Any configuration type (CV, VCV, CVVC, ARPA etc) is fine. If I'm not familiar with the configuration type you request, I will do my best to familiarize myself with it.

I can oto any language as long as a base oto is provided.
I will only oto without a base oto for any Japanese configuration, English ARPAsing and German CVCV, as I have oto'd these before and know how they work.

Diffsinger

I do Diffsinger labeling exclusively in Japanese and English, as Diffsinger labeling requires you to have knowledge on which symbols make which sound.

I'm actually still trying to figure out a payment method with low fees that doesn't flash you my full legal name so for now I only accept Amazon NL giftcards. Thank you for understanding!

If you have issues paying the full amount in one go we can discuss installments. You will not receive the oto/lab until the payment is complete.

Please reach out to me at [email protected] or @shirukagiri on Discord with the commission details before sending payment. Payment sent without commission details will be considered a donation.

Do not claim the configuration is made by you. Credit me when distributing the UTAUloid or Diffsinger.
You can do so in their readme, on their website or wiki page and/or in the description of the demonstration video.
Editing the configuration after recieving it is allowed, but please credit me for the original configuration.

Do not use the configuration for any illegal or harmful purposes such as hate or harassment.

Commercial usage of the voicebank using the configuration is allowed, I do not charge an extra fee for this.
Please do not sell the configuration I made, whether you have edited it or not.

I can cancel the commission at any time for whatever reason. If this is the case, I will refund you if the fault is mine.

1. Gather your recordings, the base oto (for UTAU otoing, if applicable) and recording transcriptions (for Diffsinger labeling, if applicable). Calculate the minimum price using the calculators in the Pricing tab.

2. Send an e-mail to [email protected], with attached a (download link to a) .zip file of your recordings and base oto (for UTAU otoing, if applicable) and recording transcriptions (for Diffsinger labeling, if applicable). Include the price you would like to pay, this must be equal to or higher than the calculated minimum price. Feel free to also ask questions or mention any concerns.

3. Wait for me to reply, this will - at most - take two days. If I have not responded within two days, send a follow-up email or message me on Discord @shirukagiri, the email most likely went to spam. If you had any questions or concerns, I will answer those first. When everything is sorted out, I will start working once you've paid.

4. Wait for me to finish. The estimated working time is dependant on the voicebank, I will include my estimation of when I'll be finished in an email to you. If it is a bigger multipitch voicebank, I will send progress updates every pitch.