Unpopular Opinions

[Prism]

I think there are many reasons for maki English having a good reception than others. First I think the want of American English vocal synth has died down now that we have Dex, Daina, ruby, cyber diva, cyber song man, Amy, Chris and big Al we are pretty satisfied. I think the reason why so many people wanted an American English accent vocaloid was that English vocaloid had a clarity problem because English is hard and it's phonetic system and hasn't really been updated since v2 other than speciality phonemes here and there. Vocaloid handles phonetic timing with velocity that is very confusing And only works well with a language like japanese. I think people confused vocaloid's phonetic clarity problem with an accent problem and since a lot of the fandom is in American or listens to American pop music they would want English with their accent. Back to maki, I think that since she is on synth v with their newer phonetic and timing systems that people are okay with accents because they can easily edit the phonemes if she isn't the most understandable. I think other factors like Solaris and Eleanor are coming out and that her ai has really nice tone helps too.

 

[WyndReed]

Avanna’s hair color has been bothering me for years. For some reason my eyes perceive that it looks strange with the rest of her design. I’ve been headcanon-ing her as having dark brown hair for so long, that it was strange to look at her illustration and see dark purple/maroon hair.

 

[Scarlet Illusion]

I actually like "moe" tuned Luka so long as it's not overdone. I think she sounds quite pleasant in Dreaming Chuchu! It's interesting how people can make a more mature sounding vocal sound much sweeter and younger!

 

[Leon]

I will die on this hill, that the default dictionary of English Vocaloids was the worst part of the banks (audio quality aside, because nearly every Engloid -- yes, DDR and my beloved V2s included -- is not great, in comparison to pretty much any other voicebank). Because the dictionary makes them pronounce words in the technically correct way, as opposed to how people actually speak and sing, they sound stilted and unnatural, and it's difficult to parse what they're saying without subtitles. And this is what I think really contributed to killing a lot of the early interest in English voicebanks tbh. If you edit the phonemes extensively (and, unfortunately, find out the unique quirks to each and every voicebank) then you can achieve good results. But this is an extra effort that has to be put in on top of the regular tuning and mixing, which comparatively makes Engloids harder (or at least, more time-consuming) to achieve natural-sounding results with.

Realistically, I don't think Maki's pronunciation is probably any better than, say, Miku's. But because A. afaik, AHS as a company and Maki as a character aren't previously that associated with English, so super 100% perfect English weren't really expected, and B. she's on SynthV, which has a more natural-sounding dictionary and is easy to edit to boot, she gets away with it without all the complaints. And I personally like her a lot too, from what I've heard!

Edit: I lost my point somewhere in all of this I think. Tl;dr, English synths don't have to be perfect, but imo they should at least be understandable to some extent. The default Vocaloid dictionary meant that the early habit of plug-n-plays were pretty much incomprehensible. I think people putting more time into phoneme edits for a natural sound helped chase off that idea that English banks just always sounded bad and needed subtitles to understand them at all. And SynthV's more natural dictionary helps accented banks be more fluid and understandable, as opposed to just totally incomprehensible, which has in turn helped lessen the amount of utter bitching that vocalsynth fans like to do

 

[Blue Of Mind]

I actually like "moe" tuned Luka so long as it's not overdone. I think she sounds quite pleasant in Dreaming Chuchu! It's interesting how people can make a more mature sounding vocal sound much sweeter and younger!

I'm not opposed to "moe" tuning for mature banks (one of my favourite Flower songs is 'Frost', and she sounds almost like a little kid), but it depends how it's done (there's a huge difference between cutesy and grating), and there are some banks that you just can't pull off the moe sound with.

 

[___]

I dunno why people still keep proclaiming how good it is that Crypton moved away from Yamaha cuz now they can update Miku and the engine whenever when it's only shown to have tainted their image, making them look a lot more unprofessional and less competent and also lessening the product's chances at being successful due to bad 1st impression ( also it looks to me like they're struggling in the development department more than ever), and also the very fact means little in the grand scheme of things when the products will always be a downgrade to their vocaloid counterparts goes for other products like Rei too

 

[ruffe]

I'm not entirely sure if this is unpopular, but I'm very much looking forward to the prospect of AI Zunko for CeVIO being a thing! Especially after the recent update for Yukari and Kiritan! While she'll likely never live up to her Vocaloid counterpart and I've come to accept that, I do think that CeVIO will handle her breathiness/raspiness much better than Vocaloid could. SVS seems to make softer voicebanks fuzzier rather than airy, which I don't particularly find pleasant. And while CeVIO itself has engine noise I think it's much more forgiving to softer voices, whereas SVS' engine tends to prefer stronger voices.

I guess it's an unpopular thing to be rooting for CeVIO too, but I believe they'll just keep getting better as the AI technology is improved and they find their groove.

 

[___]

I'm not entirely sure if this is unpopular, but I'm very much looking forward to the prospect of AI Zunko for CeVIO being a thing! Especially after the recent update for Yukari and Kiritan! While she'll likely never live up to her Vocaloid counterpart and I've come to accept that, I do think that CeVIO will handle her breathiness/raspiness much better than Vocaloid could. SVS seems to make softer voicebanks fuzzier rather than airy, which I don't particularly find pleasant. And while CeVIO itself has engine noise I think it's much more forgiving to softer voices, whereas SVS' engine tends to prefer stronger voices.

I guess it's an unpopular thing to be rooting for CeVIO too, but I believe they'll just keep getting better as the AI technology is improved and they find their groove.

I agree!

Despite the noises both of them have I do prefer how they do AI, svs' AI always comes out little stale hgvfcd Maki is all around kinda MQ/LQ release and I'm little sad her best vb ( jp AI ) sounds very little like her:/

 

[Prism]

I think maki sounds so "mid quality" is because she's still lite and from the samples when they showed off gen 3 she'll sound way better in the final release. I think updates are good thing and I think how synth v and cevio are doing a good job with them. I do wish there were samples showing of the original vs update for cevio. I think the problem with Crypton is that it doesn't look or sound finish and they keep pushing out these minor updates releasing the full boxed product with only one of her banks with the other coming "soon". I'm not downing on Crypton and I realize it's very hard to create a synth from scratch especially in a plague but it does seem unfinished and still in alpha.

 

[___]

I've always thought Maki sounded rather MQ even from the demos using her full vbs, the lite vbs only solidified that opinion and I didn't base it solely on them, I still love her and she has my favorite voicetype though.

True, and that isn't a good look for them, hence I disagree that them parting from Yamaha was good for the products' health, I think it's only been the opposite. That also doesn't mean I want NT to do badly ( ngl I'm still waiting for miss Meiko NT to drop cuz she has lovely tone hehe ), I still root for them but it was bad decision on their end regardless.

 

[cafenurse]

I dunno why people still keep proclaiming how good it is that Crypton moved away from Yamaha cuz now they can update Miku and the engine whenever when it's only shown to have tainted their image, making them look a lot more unprofessional and less competent and also lessening the product's chances at being successful due to bad 1st impression ( also it looks to me like they're struggling in the development department more than ever), and also the very fact means little in the grand scheme of things when the products will always be a downgrade to their vocaloid counterparts goes for other products like Rei too

ya know, i havent been paying that much to the piapro shenanigans so idk exactly whats going on with crypton but this got me thinking

People all the time talk about how Yamaha wouldn't let companies update vocals after they got released like it's a BAD thing, oh Yamaha is so controlling and won't let companies fix/change their product after it's been released if there's an issue, but doesn't that...make sense?
Think of it like video games. In the "olden days", you went to the store and bought your video game on a cartridge and plugged it in and that cartridge would stay the same forever, companies couldn't change it or edit it. So companies had to make that video game the very best quality and as possible glitch and bug free as they could on the first try. But now with technology advancements, companies can just put out video games in whatever state they want and update it along the way. This has been cool for some video games that were released fantastic and only updated to get even better (Stardew Valley), but more often than not, the case is the companies rush out buggy, unfinished games, lacking features and overall being unimpressive, and they say "don't worry, we'll just patch it along the way" (No Man's Sky, The Sims 4, honestly I think most EA games fit this bill?) Sure, the game could become something great after the updates, but because of that initial release where it totally sucked and was a giant mess (No Man's Sky) it forever has this tarnished reputation. People that just gave it an initial try and were like "oh this is garbage" aren't going to want to go back because it's better now. And diehard fans of a franchise are made they had to wait YEARS for a game to be updated with the content it should have had from release (grumble grumble Sims 4)
Voice synths can be treated the same way. Company puts out voicebank. Voicebank sucks. It has technical and quality issues. People spend a lot of money on this voicebank, try it out, and go "wow this is trash". Months later, the company goes and patches up said voicebank, sends out the update "look it's good now!"
But shouldn't it have been good from the start? Why did these people that paid full price for this product have to wait for the company to go back and fix it? It should've been fine when they bought it.

So personally, when everyone scoffs at Yamaha for not allowing companies to go back and update their vocals, and praise companies for breaking away because now they can, I kind of see Yamaha's point of view. Like Patuk said, it comes across and unprofessional, less competent, and for people that try out the product when it's released, they have a negative 1st impression that's hard to shake.

 

[ikaroll]

But now with technology advancements, companies can just put out video games in whatever state they want and update it along the way.

Ya know, i was trying to figure out why Piapro NT seemed a bit iffy to me, and that just summed up WHY
I feel like they should focus on Miku and Miku only. Yes, its nice to have some updates on the others, but its been *checks smudged writing on arm* 7 months since Miku NT released, and I just think it would have been a better product if they had just focused on finishing only Miku before starting on the other Cryptonloids

 

[___]

So personally, when everyone scoffs at Yamaha for not allowing companies to go back and update their vocals, and praise companies for breaking away because now they can, I kind of see Yamaha's point of view. Like Patuk said, it comes across and unprofessional, less competent, and for people that try out the product when it's released, they have a negative 1st impression that's hard to shake.

Thank you I was getting exactly at that! It might seem strict at 1st but I do think Yamaha was onto something, in this climate voicebanks get picked up mostly in their 1st month of their release and people usually don't go back to them and 1st impression is the end all in this industry, there's reason why most vocaloid updates are just straight up ports.

I do think some updates are inevitable in the AI era as it's improving at a fast pace ( unlike sample based technology which had been stagnant for a hot minute ) but that's outside the voicebank creator's control but that isn't an excuse to make poor quality AI vbs /all shade Yukari Rei.

I am the biggest V5 stan but it fell victim to the same thing when Yamaha started being more loose with it's updates in this era of vocaloid, it's miles better editor than it was in 2018 but it's still given slack.

 

[Chuchu]

I can't seem to warm up to Kotonoha Akane & Aoi. They sound okay, but everytime I hear them sing I get bored real quickly lol

---------------

I actually think it was good for Crypton to part ways with Yamaha soon enough before their business relation sours beyond repair. Crypton should just not have announced Miku NT's release date too soon and they wouldn't be in such a mess right now. I plan to buy Luka NT whenever she drops and am actually relieved that, even though most are minor patches, Crypton is willing to fix their products instead of letting them crash and burn.

 

[blackout]

So personally, when everyone scoffs at Yamaha for not allowing companies to go back and update their vocals, and praise companies for breaking away because now they can, I kind of see Yamaha's point of view. Like Patuk said, it comes across and unprofessional, less competent, and for people that try out the product when it's released, they have a negative 1st impression that's hard to shake.

I mean... Sure, but this still absolutely did not stop them from releasing some doo doo sounding LQ commercially released Vocaloid banks. The ability to provide voicebank updates doesn't mean that quality control goes out of the window entirely...

 

[Chuchu]

I feel like they should focus on Miku and Miku only. Yes, its nice to have some updates on the others, but its been *checks smudged writing on arm* 7 months since Miku NT released, and I just think it would have been a better product if they had just focused on finishing only Miku before starting on the other Cryptonloids

I think Project Sekai is partly responsible lol Craftegg seems to like to have a balance between the Cryptonloids than solely bank on Miku, which force Crypton to divide their attention between their products more than usual.

 

[___]

I mean... Sure, but this still absolutely did not stop them from releasing some doo doo sounding LQ commercially released Vocaloid banks. The ability to provide voicebank updates doesn't mean that quality control goes out of the window entirely...

Yeah but they're few and far between and we already saw how this mindset lowered the quality standard in games.

 

[ruffe]

The only LQ voicebanks I can think of right off the bat are either V2 banks (SF-A2 Miki, Sonika, Lily, The Kagamines), or are the Meikas & Miku CHN. After the V2 era most/all banks released were of medium to high quality. Of course, Voicebank updates don't mean quality control flies out the window, but it does mean that there's less effort to make a great bank right off the bat, which can leave a sour taste in some peoples' mouths.

Who knows, maybe VOCALOID has spoiled me in that I'm used to expecting HQ products... I mean, the V3 era didn't have a single notable LQ release to my knowledge, nor did V4 (except for Miki's repackage and Arsloid's CV banks, of course) haha.

 

[Scarlet Illusion]

Although I really love CFM's products thus far and am looking forward to seeing further developments with NT, I think they should have waited until Miku NT was actually completed before releasing her and before working on the other NT banks for ProSeka (is there some reason they couldn't just use their already existing and complete Vocaloid banks instead of an unfinished product..?). Also instead of giving estimated release dates for updates and so forth, they should just keep quiet and mention update releases when the update is actually complete. :S
Or, if they're going to want to work on the others at the same time, hire more people and divide into different development groups for each synth? (I realize I'm oversimplifying a bit, but surely with everything they make off of their vocas, Miku and co. merchandise, etc.. they'd be able to hire and pay people to assist in development??)

The only LQ voicebanks I can think of right off the bat are either V2 banks (SF-A2 Miki, Sonika, Lily, The Kagamines), or are the Meikas & Miku CHN. After the V2 era most/all banks released were of medium to high quality. Of course, Voicebank updates don't mean quality control flies out the window, but it does mean that there's less effort to make a great bank right off the bat, which can leave a sour taste in some peoples' mouths.

I'm surprised to hear you say the Meikas are LQ! In my experience I would've placed them to be around more MQ--the only thing I've noticed that I find genuinely bothersome about them is that Mikoto can sound a bit stopped up/stuffy nosed/singing like she has a sock stuck in her mouth at times. Hime's a bit breathy but it never detracted much for me! I've never really found them to be difficult to work with, and shockingly enjoyed working with them more than any Vocaloid I own aside from Gumi!! Mikoto (I can't vouch for Hime as I've yet to attempt this with her) is about medium difficulty to mix as well.
I'm curious about what you have to say about them!! Spill that tea

Also, an unpopular opinion to keep us on track:
I think SynthV as an editor is good, but its engine is needing improvements. I can't quite place my finger on what is about the engine that's a turn-off, but I feel like all the voices sound kind of... stiff, I guess? It surprises me considering the developer, Kanru Hua, also made Moresampler which is the UTAU resampler I swear by.

 

[ruffe]

I'm curious about what you have to say about them!! Spill that tea

Okay so! I may have been a little harsh in calling them both LQ since Mikoto is definitely more MQ than anything, and I share the same complaints that you do. However, Hime is a different story for me... The best way I could describe her would be as if someone turned CLE all the way up while using an airy bank! It makes her a bit of a pain to mix/rectify, and when you compare her raw output to another airy voice like Tohoku Zunko or Aoki Lapis it very much shows her problems!

that said, she's definitely good in the right hands, just like any Vocaloid is!! But judged purely on her raw output I'd say she isn't too great of a loid in terms of quality.