#SetParam

[vok4l0id]

Download SetParam & Oremo

Unfortunately,the official pages for Oremo and SetParam aren't working correctly but I have the files of the programs uploaded on MediaFire so you can get them from there

-SetParam

-Oremo

(You will need WinRAR to install them,you can get it here)

[babbybones]

so i've been thinking... maybe undertale fans should be more open to checking out UT AUs! i know some of us might be inclined to think they're not as good as "the real thing," but there's really a rich, creative community behind them. not only that, but it's fun to make your own too! here are some of my own recommendations:

you can't go wrong with the old classic Tachimukae! Kimi wa Kakkoii! (Face It! You're Cool!) the animation is a love letter to the worldwide UTAU community of the time (2009) and features cameos of practically every UTAU that existed back then, including those made by young amateurs overseas. this was a Big Deal when it dropped.

i still really love the original song null by fractalsleuth:

and if you were wondering, UTAU isn't limited to japanese. english voicebanks are possible too!

you can get started with UTAU by installing either the open source, multi-platform OpenUtau, or by switching your locale to Japanese and installing the original UTAU for Windows. i'm personally used to UTAU, but OpenUtau is in active development so i think it's worth trying out :)

you can think of a very simple UTAU voicebank as a folder containing .wav files of Japanese syllable sounds, like か.wav (ka) and so on. so just get your nicest microphone, a reclist (recording list), and start recording! OREMO and SetParam are highly recommended software for recording and configuring UTAU, as they're made specifically for it.

you'll find many ways to record an UTAU, but a CV (consonant-vowel) reclist is definitely my recommendation for an absolute beginner. even Wayne can do it...

[modsyvoxynth]

Hi Tumblr :)

I remember this cover being very fun to work on. I had just downloaded Oremo and wanted to test it out, so I recorded this weird, raspy VCV. Seriously was blown away by how easy it made recording voicebanks and I believe I also had just downloaded SetParam. Nowadays I use vLabeler for OTOing and Ive been wanting to try out RecStar :)

[roseverdict]

fuck it. bill cipher utau incoming whenever i finish figuring out how setparam works

[zipzapproductions]

Most vocal synth companies these days are like “Here’s an ai!”

But with us you’re getting “Here’s an arcade operator with 30 samples and a shitty setparam oto”

[star-thorns]

setParam my beloved. Otoing a VCV bank with like 200 wavs and 1000 lines of oto my beloathed

[ensoua]

utau recording is just opening setparam. closing setparam. editing the oto. open setparam. close setparam. edit the oto more because setparam didnt recognize what youre trying to do. opening setparam. and so on,

[roburohaj]

hey, rook here

while were at it, heres a screenshot of setparam.

i have 4 pitches recorded, and im trying to do a few more. buros english voicebank isnt even finished, either. ill probably release the voice color prototype version with their monopitch SOFT and DARK english expressions so that i can focus on this new bank while not neglecting rk-64 buro jaguar (english bank). ive been meaning to make a few of 32x's (cv-vv bank) expressions bi-pitch, but i dont think thatll be all too possible, sadly :[

ive been considering the idea of a talqu or coeiroink talk bank, since unicoe just cant seem to work on my pc, but dont count on it, the documentation made my head hurt xP

[uta-uta-utau]

hi i explained all of my plugins (hopefully well enough) in one video.

[nitw]

someone should take one for the team and make a reigen utau vb so we can make him sing toxic

[skelevision]

learned to use setparam! unfortunately i still hate otoing

[willow-dot-file]

Oto.ini is pretty fun to set up ngl.

Except vowel samples in VCV, fuck vowel samples.

[raytaku]

I’m pretty sure that my least favorite thing to do utau-wise is otoing 

[utau-4-you]

VCCV English Tutorials and Resources

VCCV is a recording method for UTAU English created by Cz and released in 2015. VCCV is considered one of the community's standard methods for creating English voicebanks.

PROS AND CONS OF USING VCCV

CONS:

VCCV takes a while to record and OTO (4 hours of recording per pitch) and creating USTs can take some time, making it a daunting task to start using. As well, the amount of recordings needed can make it harder to create multipitch or voice acted vocals. It's recommended to use a method such as arpasing or delta's English list if these are your concerns.

PROS:

VCCV is a modular reclist meaning that it's incredibly easy to add or take out features. Because the recordings are standardized per each modular/folder, it's also easy to add custom phonemes. Unlike arpasing's word based reclists, phonemes are consistent across the whole bank and don't require duplicates. VCCV also has a large amount of distributed USTs and resources available to use. VCCV is a very accessible method due to there being a very in depth series of tutorials written by Cz, the reclist's creator.

TUTORIALS

Cz has gone through the trouble of creating video tutorials on how to record, OTO, and use VCCV. It is recommended that you watch all the videos before you start using VCCV!

There is also a text transcript of the tutorials that you can view here

You should use OREMO and the provided recording BGM for recording a VCCV bank as it will aid you greatly when it comes to the OTO process.

It's recommended that you use SetParam to OTO as it will speed up the process. Cz goes over how to set up SetParam in the OTOing portion of the tutorial. You can also use the built in OTOing tool in UTAU or vLabeler, an open-source alternative to SetParam. It doesn't matter which you use as long as you follow the techniques shown by Cz!

RESOURCES

VCCV English Phonemes Guide: A google spreadsheet of all the phonemes for VCCV, pronunciations and x-sampa equivalents. Includes links to recorded samples for most of the phonemes.

English Phonemes Cheat Sheet: A PDF of all the phonemes for VCCV with their Vocaloid x-sampa, Delta, Arpasing and CV-C conversions

Lyric Parser 2.0: A plugin for UTAU that converts English lyrics to VCCV English. The dictionary for the plugin is incomplete so it will not 100% convert everything you give it. You will have to fine tune what it does convert as well. However, it is a great start and saves a lot of time when it comes to UST creation!

If you are using OpenUTAU, the program does have a phonemizer for VCCV English that is rather good and takes away most of the struggles of using a VCCV bank (though you should fine-tune it to the voicebank you are using.)

[utau-bowl]

Yui Yuuka Review

Hello! After a long time, I finally have another review that I did for fun. Downloading random UTAUs isn’t something I do much anymore, so I rely on requests, my friends, and social media to show me UTAUs I don’t know of. Yui was assigned to me during UtaPlanet and I liked her voice so much I wanted to talk about it more.

*Art by Deco.

Hear a Sample of Yuuka Yui

Bio

Name:  遊歌ユイ / Yuuka Yui Age: 13 Height: 4′10″ Weight: 90 lbs Nationality/Race: Japanese Voice Actor: shizy/Deco

Yui is a happy-go-lucky junior high student who dreams of being a professional singer. She likes video games and magical girls.

Official Site

Yui doesn’t have a proper homepage or website, but does have a couple wiki pages. Her download is hosted on Bowlroll (password is “yui”). Her voicebank is 407 MB when unzipped, and has fully generated frqs along with partial llsm and pmk files.

First Impressions

Yui is a soft, childish voice with a noticeable lisp. She has a westernized accent since her voice actor is from Canada. It seems her creator is inactive in the UTAU community at present.

Configuration

Yui is a tripitch VCV recorded at C4, G4, C5. She includes glottal stops and vocal fry for all vowel combinations. There are some "extra” sounds like tyu, di, wi, ng, and wu. All the o samples have alternate aliases with を. Overall, the reclist is quite minimal, with just 170 samples. I found that her default vowel endings sounded kind of strange so I ended up using the CV breath samples a lot. The default ones just take the end of an existing string, and by CV breaths, I mean ones that were recorded separately.

The volume and mic quality is consistent overall, making for a clean sound. There are no mic pops or background noises that I found while looking through the recordings. As is expected with the pitches she’s recorded at, she has a very flexible high range. Her C5 pitch actually has a lot of energy, which I love. It’s not really a powerscale, but the lower pitches are softer.

Most of the oto uses whole numbers, with only some entries having the trademark setparam decimals. The overlap of nearly every entry is 100, with the ratio of overlap to preutterance being a consistent 1:3, occasionally 1:2.5.

My recommended flags: F2B0Y0H0C99

My recommended resamplers: doppeltler, world4utau

Final Thoughts

I didn’t have a lot to say about Yui. That’s because her voicebank is very well made, and there’s only so much I could say about her voice type. This is the only voicebank for her the creator ever released, and it works well as a standalone. If they ever came back I would love to see a more powerful Yui voicebank, like the C5 pitch made into an entire bank.

Got any other UTAUs you want me to review? Send an ask!  

[tananansad]

“what got you into linguistics”

Asked by @asyirafu​ (Lost your question in the inbox, sorry!)

I used to have this small, black ring-bound hardcover notebook with a smiley on the front cover where I'd diligently write down words from this one language-learning website for various languages, basically becoming my phrasebook; I was so obsessed with the idea of being a polyglot sometime around kindergarten or lower primary. Of course, the only things I remember from that phase are a good chunk of Japanese and auf Wiedersehen.

That isn't really the answer to the question, but it does mark when I realized I has a certain liking towards the concept of languages. What got me into linguistics however...

... that requires a lot more explanation than some kid’s jot-downs.

I found myself caught in what was essentially a niche storm. Initially I was really just fond of the uncanny value these singing robots had to offer, but I soon decided it was time for me to dig in a little about the technical aspects of artificial vocalization. Knowing that Vocaloid is paid software; I had to go for a free alternative.

Squiggles galore! It took me a good few years and endless hours of isolation to perfect the nerdling skill of “tuning” (i.e. detailed vocal manipulation); I still wouldn’t say I’m the best at it, however. [Image source]

Naturally, I picked up a few things about concatenative vocal synthesis, identifying phoneme onset timing via waveforms, spectograms and a lot about prosody!

The natural habitat of an UTAU voicebank creator, SetParam. [Image source]

What’s more important, however, is how UTAU introduced me (and a LOT of manual vocal synthesizers in general) to... linguistics! (Woo, finally onto the main point!) Heads up, lots of savvy examples up ahead!

Reclist makers MUST have a full understanding of phonological processes that occur in their target language along with a good whiff of dialectology (esp. for English) to create a good reclist that allows synthesized vocals to sound as natural as possible. Phonemic holes in a reclist can result in forced concatenation alternatives which will sound unnatural. This learning process took not only UTAU users but also Vocaloid and companies working on English Vocaloiods years to learn.

For UTAU reclists, specifically, the phonemic transcription system used is usually meant to be intuitive to a native speaker. (Vocaloids jkust use a modified form of X-SAMPA.) Japanese reclists usually stick to kana, but because of the inconsistency of the English orthography, many systems have been created, the most pervasive one being VCCV.

So yes; singing robots got me into linguistics, and I’d say that’s pretty rad!

With all that being said, UTAU, voicebanks and reclisting weren’t necessarily the only things that influenced my affinity towards linguistics. (Asyraf’s went through my blog, obviously but) If you’ve been through my blog, you’d know that I’m a heritage speaker of a language that’s slowly dying. On top of that, most of the media I’ve grown up with are a hodgepodge of languages, so I was basically raised to be an acute listener to my multilingual environment.

Hoped that answered your question!

Notes:

A reclist is a ”recording list” one uses as a guide to recording a full voicebank set; also usually includes “oto”ing guides and the like.

SetParam is a program used by UTAU voicebank creators to record lines from a reclist in tempo and configure each recorded line to form separate chunks of syllables ready to be concatenated in the UTAU program, a process known colloquially as otoing, from the name of the .ini file one’s voicebank’s configurations are saved in.