#ARPAsing

[dead-byte]

So, since getting rid of Twitter Circles is apparently going to be among the Muskrat's poor decisions, I'm going to start releasing some of my more complete experimental reclists that I've kept exclusively to Twitter Circles.

Starting with this one:

This is a simple phoneme-based ARPAsing list I've made. It's a 3-mora list, and contains only standard ARPAsing phonemes ( + an optional "dx" addon list ) with full diphone coverage.

It has "index" and "no index" variants ( the "index" list is recommended if using moresampler to generate a base, due to sample priority ). While is has no extra phonemes, it is meant to be friendly for non-rhotic accents as well, as there should be enough instances of r-colored vowels to act as contextual duplicates.

In addition, it also comes with comment files that transcribe ARPAsing phonemes to SAMPA(DELTA-style), VOCALOID SAMPA, and VCCV, for those more familiar with other formats.

There is also an IPA Transcript, primarily meant to act as an alternate script for more experienced singers, who may be familiar with IPA, but none of its' adjacent phoneme formats, or SVS in general.

The default list ( without the dx addon ) is 516 samples. However, there is an even shorter ( optional ) reclist for additional pitches as well, that cuts out a lot of sounds that don't change much recorded at other pitches. The idea being the base pitch layer of a voicebank uses the "default" list, and the "additional pitches" list is used for the rest of them. The "additional pitches" list is 345 samples. That said, you can use the default list for all pitches too.

This list is untested, so there may be mistakes. However, it's a fairly simple list, so I don't anticipate any huge errors. That said, please do let me know if you find any.

[dailyenglishvoca]

Today's song is a string of coincidences by Du Du Danyon featuring the UTAU voicebank Hexia

[nana-yoyo]

The original song

[utau-bowl]

New Thoma just dropped

[gretahaffandhawf]

I'm finally starting to get the hang of this. Kudos to cosmicdorito for developing CVVCymraeg!!

[tomatetoro]

a test of haru ai on sovits! he's at 24k steps,, he's able to stretch out longer notes now which is rlly cool >:3 i think i wanna train him more tho in my freetime

[cosmindart]

I just wanted to shill WavSEED's 4 VBs again because I finally made him a wixsite

(I still prefer his Toyhouse page tho XD)

Ecume

Tide

Tsunami

Ripple

[henb-y]

also cuz I yelled abt it on twitter I'm yappin abt it here too. get RIKU'D

[rucow]

ive been recording 1 arpasing pitch for caoimhe each morning (in one sitting cuz im insane) and everything's great and all but oh my god there's so many files to oto😭 i got the hang of it but there's sooo much idk if ill ever finish it??? ill try otoing a bit each day but idk if its gonna be finished before caoimhe's first anniversary in june omg

[shiweimigi]

Shoutout - Yours To Devour

Original Song Shoutout featuring a JAE VOCAL PROJECT vocalist!

"Yours To Devour" by frutigerangel featuring Canary, Milk, and Akihana

[skelevision]

mm i wanna keep working on that jinriki bank slash string theocracy cover but ive started making usts by makign a midi first and i just got a plugin that can split notes so thats probably a better method than what i Was doing. but that means redoing all that hard work...... ive only done like 6 lines so far though so i shouldnt be so hung up on overhauling LMAO. there is Much More of the song i should be worried about

actually wait i do have like a few more lines than i did last time i showed this

[dead-byte]

Sooo... some of you might not be aware, but Mine Laru received a new voicebank like a week or two ago ( a whooping 8 pitches, jesus, not even a lot of native voicebanks are that big ), and I have a lot of ( mostly positive ) thoughts on it, along with the new ARPAsing reclist made with it. I wanted to compare it with an older Mine Laru bank in my possession, thus the audio clip above. You can find the new bank here:

Now, some of you might know of the new AgO voicebank, but not know of the 2020 voicebank. Let me explain that one. There’s a bit of a story here.

A couple of years ago, Minato ( Mine Laru’s VP ) set up a website called the “Mine Laru Lab”, and it was to help test out and give feedback for a lot of new voicebanks that were made for various languages that weren’t Japanese. Among them, was a dipitch English ARPAsing voicebank based on the 0.2.0 list.

I offered to configure it in a Twitter comment, and Minato seemed to welcome my help, so I did that as soon as the site went up. After I finished, I DM’d Minato, letting him know I’d finished, and that I had suggestions. I never received a response, and in retrospect, I’m thinking he probably didn’t have his DMs open and I simply never realized. I never followed up though, because I didn’t want to bother him.

Sometime around then, the Mine Laru Lab was shut down, and access to the various voicebanks was seemingly discontinued.

I could be wrong about this, but I think this might be the voicebank that was once called “Mine Laru PRIMITIVE”. I don’t know that for certain, so that’s merely a guess.

So… I’ve basically just had this obscure completed English voicebank sitting around on my computer for the past couple of years. I’m not going to distribute it or anything, unless I get explicit permission to do so. I figure showing it off is probably okay based on what I remember of the Lab TOS, but I will delete this post if Minato asks me to, on the off chance that happens.

Anyways, the AgO voicebank came out recently, and it sounds pretty different from this older bank, so I wanted to make a more direct comparison. The 2020 voicebank is a lot more similar to Mine Laru’s more powerful voicebanks, such as his 14-scales or his “query” voicebank. The AgO voicebank sounds more like his “parse” voicebank, with a more neutral tone becoming softer in higher ranges.

I also want to talk about the reclist, because in my opinion… the AgO reclist is kind of a stroke of genius. It’s such a simple idea that I’m embarrassed I didn’t think of it first. An easy word-based list based mostly on singular words, most of which being easy to pronounce for non-native English speakers. That’s just… sort of brilliant.

And this might be controversial to say, but… it’s leagues better than the 0.2.0 list. The AgO list has just about every CV, VC, and VV diphone you could need for English. There are some exceptions, it’s not 100% complete, and it is missing a lot of CC diphones, but it’s so much more useable than 0.2.0. Reading the blog post on the process of making the list is extremely fascinating too.

I also made an experimental dictionary for AgO to hopefully rectify some of the diphones that are missing. Again, mostly CCs. This dictionary isn’t anywhere near as chaotic as my 0.2.0 dictionary thankfully. Though, there are still some rather obtuse transcriptions I had to use to get around the list’s limitations.

As for the voicebank itself - Mine Laru has always had good English. Even his old “multilingual normal and weak” voicebanks were really good. Mine Laru in English is the kind of voice where he has a notable accent, but in a good way, like how Megurine Luka and Tsurumaki Maki do, where the accent adds charm and personality to the voice instead of making it more difficult to use. The AgO voicebank is no different.

In contrast to the 2020 bank in particular though, having gone through a lot of those old samples, it’s clear that the VP struggled with the 0.2.0 list. Most people do after all, even native speakers. For all its simplicity and compactness, I cannot tell you how many times I’ve gotten a commission to configure a 0.2.0 voicebank, and the voice provider audibly has trouble in the samples - and to be clear, that’s not the voice provider's fault, 0.2.0 is just a bad list ( I’m looking at you “yamk”; that’s not even a word ).

In the AgO voicebank however, the pronunciation is so good. Like, his pronunciation is good to begin with, but with the AgO voicebank, his pronunciation sounds a lot more relaxed and comfortable. And even better, the AgO voicebank actually distinguishes between aspirated and unaspirated plosives via alternates. By default, most plosives are unaspirated, but if you cycle through its first alternate ( by typing "1" at the end of the diphone ), you get an aspirated consonant. This is the kind of weirdly specific English voicebank construction that makes me absolutely giddy. Clearly a lot of thought and care was put into this voicebank.

My only real criticism is… the oto.ini. It’s got some problems that can lead to the voicebank being rather choppy. A lot of start-phrase consonants and CC diphones have problems too.

I kinda… want to stick the voicebank in vLabeler and try to refine the oto.ini some. I might do that.

Anyways, I just wanted to dump my thoughts, in case anyone wanted to listen.

[balzabul]

original song by niki english lyrics by akiglancy art by ROYALoid (me) ust by WinterdrivE

My teenage dream come true, an English UTAU voicebank for Hauspie! Voice source is the default male voice from Speakonia; he's been "Hauspie" to me, my own little OC, for over a decade now, and I hope you like him! He's not a singer by trade (lol) so his design doesn't reflect one.

This was also my first arpasing bank?! Go figure.

[utau-bowl]

A cover of Amitai’s “Patron of the Plague” with Kayaka Konagi. A UST download is linked in the description.

[lesbian-forte]

I wanna say a few things about Synthesizer V, because a lot of vocal synth fans, especially of mainly Vocaloid and UTAU, tend to misconstrue intent when it comes to wishing for vocals to migrate or cross-platform on the engine.

-It isn't about 'realistic,' or at least for the most part. It isn't about disliking their classic robotic sound just because of progress. I'm gonna say it- I do dislike Vocaloid 6. I don't think that's an unpopular opinion. But I don't hate Vocaloid in general, and want badly for the software to improve.

Banks like Gumi and Una got smoother transitions and (bad) crosslang... at the cost of getting put through a white noise factory. Their V4s were much clearer and even Gumi V3 English was far more reliable, while their V6 updates sound as though someone in the recording booth is rubbing styrofoam and the mic lacks a pop filter. Supposedly in the name of the iconic 'Vocaloid sound' that the type of engine noise doesn't even resemble. And for those who like the robotic sound and don't like SynthV because it's not, I've got news for you- you can have it without the fuzz.

For one, you can tune robotically in SynthV very easily. Tune entirely manually like any non-AI program, and since there's very little of that metallic twang, certain effects just move to mixing. Rather than taking out the sound of engine noise, it's putting certain things back in instead, which is a lot easier to do than the inverse- and the option of keeping things sounding more natural and clean without all that work is a huge bonus for people who don't think that sound fits their taste or style.

I personally love robotic-sounding vocals. But a clean, clear, and fluent base render output is essential to get the most out of your sound. I love Eleanor Forte R1/lite because that's what she is. Clean, clear, and fluent. But not realistic in any sense, unless compared to previous vsynths. She's probably just an Arpasing bank with higher render quality, but she's amazing regardless. And free too!

-Accessibility. SynthV is very accessible and easy to pick up. There's a limited version to try it out and make projects at the cost of fancy features that even allows commercial use, and many voicebanks have lite versions also with limited features and no commercial use. There aren't time limits on trying out the program- and just like Vocaloid, if you tune something with one vocal you can send it to a friend to render with another. Having lite voices to work with a lower quality version of the same voice that will render it means you actually know for the most part what you'll get, save for a couple quite old exceptions.

And when you do want full versions, it's dirt cheap. It's cheaper than Cevio, and far moreso than Vocaloid or Piapro. The two most expensive voicebanks on the program convert to roughly $120USD, and they're known for their special features and high quality. Most are 60-90 for digital, and SynthV voicebanks are feature-rich with the equivalent to appends already built in.

Full versions of AI voicebanks (the current gen) are also multilingual with little quality loss between languages, mainly just an accent. No need to purchase a separate English voicebank with no appends or deal with extremely clunky input and tuning. Currently four- (soon five!!) are options usable for any voice. Japanese, Mandarin, Cantonese, English, and when testing is done, Spanish will go from being exclusive to one bank to fully available.

It's also just generally very user-friendly. Especially if you dip your toes in with the basic editor before being barraged with all the features of pro, everything is laid out simply. While the UI could use a minor revamp because it's getting crowded and being able to resize the sidebars would be nice, there isn't really anything bad to say about it. As someone who'd hardly touched vocal synths myself, I knew how to work it in hours.

Oh, the autopitch. This one is contentious. There is a certain laziness to doing plug-n-play on a cover and calling it a day, just like with any other vocal synth. However, on originals, or working with it to make sure the voice cooperates, more time can be spent on phonemes, parameters, and mixing. Calling SynthV bad because it's just 'too easy' is gatekeeping vocal synths, plain and simple. Being able to make what you want without having to manually make every single pitchbend or fight with the program before getting to do anything more interesting is a way to make people more creative, not less.

-In regards to people complaining about the updates- aside from Stardust because of her limited copies and discontinuation, previous versions are still there! Dreamtonics doesn't give companies or individuals predatory contracts that force them to stop distributing older products with the same VP or wait for years to move. They're okay with their voices being on multiple engines and still supported. You'll still get your UTAU Teto, Renri, Oscar and XYY, Cevio Rose, Popy, and Kafu, and Vocaloid Gumi, Sora, Miki, and Kiyo. And you'll get your SynthVs!

Synthesizer V, Voicepeak, and Dreamtonics are not turning into a monopoly. Vocaloid still has tenure and the current version has a backlog spanning back to V3. Cevio has some heavy hitters that make it huge in Japan. Ace Studio is big in China and has the Vsingers. If you see characters coming to SynthV, that speaks of the quality of the program and its capabilities of bringing a wider audience. It has naturalness for those that want it, high quality render output, expressiveness, fluent English for marketing to the west, and Windows, Linux, and Mac support. There's nothing wrong with wanting options, and the companies know this. It's broadening the market, not vice versa.

[joezcafe]

Can you please share the Welsh reclist with me? In addition, would you consider releasing it publicly at some point, since I wasn't able to find it on your website?

Hi Greta!!!!

I'm so so SO sorry for keeping you waiting for so long, I just wanted to give the whole thing a look-over and was procrastinating on it.

Here's a download to the beta reclist with an oto base, I hope you and anyone else who's interested gets some mileage from it :D