Unpopular Opinions

sunny_likesvocaloid said:

project sekai and project diva are both good games and the comparison between a CONSOLE game and a MOBILE game is unfair (i've seen that in the vocaloid games community)

Click to expand...

I can see both sides of that discussion... a lot of us older fans are frustrated that there will be no more Diva games while Sekai is thriving. The reasoning why they won't develop another Diva game didn't made that much sense to a lot of fans and even though you're comparing console vs. mobile game you're still comparing two vsynth games that both have some characters and songs in their respective setlists in common. In fact I've seen a lot more of such comparisions than ppl comparing the actual gameplay. However I have seen Vocaloid fans that are weird about it so ig fair that you bring it up here

sunny_likesvocaloid said:

project sekai and project diva are both good games and the comparison between a CONSOLE game and a MOBILE game is unfair (i've seen that in the vocaloid games community)

Click to expand...

The idea that mobile and console games can't be compared is kinda silly... In this day and age, there are mobile games with budgets just as big or bigger than some console games. They're not all indie games or Angry Birds anymore.

Alphonse said:

The idea that mobile and console games can't be compared is kinda silly... In this day and age, there are mobile games with budgets just as big or bigger than some console games. They're not all indie games or Angry Birds anymore.

Click to expand...

eh, there's specific lmitations when making game for mobile. It's a constraint that radicallly alter the deesign consideerations of the assets (they generally need to be much more low poly in general, and certain engines and graphics effects do not work). it's still like that, so you need to change the whole style to fit smaller hardware. things are different if you don't concern 3d game and focus on 2d type animation, those can compare very closely with some exceptions. Live2d cubism probably has limitations with too many layers in a mobile game so effects/highlights can't be too complex.

How is the ACE loading time overseas though? I recall one big problem with ACE outside of its controversies was that it was cloud based with servers only in China, so it was super slow everywhere else

Leon said:

How is the ACE loading time overseas though? I recall one big problem with ACE outside of its controversies was that it was cloud based with servers only in China, so it was super slow everywhere else

Click to expand...

According to ACE's own statistics published within a DTM magazine interview (iirc), majority of the users for ACE Studio are actually from North America. From this, I'm assuming that the load times have been drastically reduced and perhaps investment in Amazon servers for more region coverage. Their article did not specify how many users they were experiencing overall though (which I find to be peculiar with the alleged claims from an ACE studio member themselves sometime at the end of last year who claimed they were barely turning a profit).

Releasing 26 AI voices is a feat, and many of them do have a high likelihood of appealing to the general producing demographic, but I'm still not too sure about the staying power of these voices and ACE as a whole. While the controversies within the fandom are definitely not known to outsiders, I think that their singer icons are actually not stock images, but AI rendered images which is still a controversial move in most online spaces nowadays. Coupled with the price of ACE (subscription) as well as being cloud-based it's hard to really gauge if this is going to pull them out of the alleged financial deficit they were in, unless they received a new investor. While subscriptions are very common for VST instruments, cloud-based + subscription is not - it's usually one or the other.

While I'm still not sure about ACE's overall power or their ability to truly seize the demographic due to the business methods itself, outside of controversies, I do agree that Dreamtonics needs to step it up. The DT vocals are across the board very good products, but there is an increasing problem as of late with quality control both in their sound quality and visual representation (the recent Cantonese female is a great example of the sound quality issue I'm mentioning. It feels like with every letter-voicebank release the overall noise, but especially on consonants such as (s) is becoming more and more noticeable). The letters are a generally good concept, but the fact they look like they were slapped together in photoshop with no care for consistency among spacing/crop or even background color (many of them do not share the same background color), it can definitely come across as cheap or lousily made.

I agree that the niches filled by Dreamtonics voices are also not nearly wide enough, and the ones that are missing are deeply felt from a contemporary production standpoint. The fact we still don't have what many might consider a "true english female pop" vocalist from Dreamtonics is one that immediately comes to mind. DT definitely needs to focus on ironing out more of the kinks within the letter releases that are getting more prominent, focus on some desperately needed genre variation that average composers/producers outside of the vsynth sphere would need, and improve the visual representation as well. (Not to mention how many voices are taobao exclusives for an insane amount of time, we still have no native Spanish letter, and there's no word on if we'll be getting a letter voice for Korean.)

lIlI said:

ACE Studio recently released 26(!) new voices specifically targetting the exact needs of professional producers: I believe they'll overtake SynthV among casual English-speaking users if Dreamtonics doesn't rapidly increase development and improve their marketing.

ACE uses visually pleasing stock images for their professionally oriented vocals, which is much more appealing to that demographic (and perhaps most people) than DT's inconsistent spiky letters. They catch the eye, provide a clear picture of what type of voice they are, and still feel 'pro'. Sometimes, the oldest vocal synth marketing tactics are the best, haha; Leon and Lola knew. ACE's controversies aren't well-known outside our community, so their reputation can't hold them back outside the fandom.

There was a period when it seemed inevitable that SynthV would become the defacto vocal synth for professionals, but Dreamtonics has lagged behind in its development of English voicebanks, leaving Eleanor, Natalie, Anri and Solaria as the only feminine options for years; which no real choices for common genres like RnB, heavier EDM, a ton of vocal trends in pop, pop punk...etc...etc.... Combine that with nonsensical ads that fail to explain the product, terrible visual design for their in-house vocals, and poor SEO making their outdated SynthV R1 appear first on Google (so many potential customers have bounced off because they accidentally downloaded the wrong program) they made it easy for competition to come in and eat their lunch.

Click to expand...

I am extremely doubtful of that considering how much the same demographic downright despises subscription based software... which is what ACE is.
Yeah I just cannot agree to this take, no offense.

edit: Small addendum, I don't have much to say about first party voices and their potential issues cause quite frankly I do not give a shit about them so I don't have much to ad. But I feel SV and by extension Dreamtonics is doing well enough where I feel they don't actually have to do anything, but that's just my two cents. At any rate some QOL inprovements would be nice I will say.

lIlI said:

ACE Studio recently released 26(!) new voices specifically targetting the exact needs of professional producers: I believe they'll overtake SynthV among casual English-speaking users if Dreamtonics doesn't rapidly increase development and improve their marketing.

ACE uses visually pleasing stock images for their professionally oriented vocals, which is much more appealing to that demographic (and perhaps most people) than DT's inconsistent spiky letters. They catch the eye, provide a clear picture of what type of voice they are, and still feel 'pro'. Sometimes, the oldest vocal synth marketing tactics are the best, haha; Leon and Lola knew. ACE's controversies aren't well-known outside our community, so their reputation can't hold them back outside the fandom.

There was a period when it seemed inevitable that SynthV would become the defacto vocal synth for professionals, but Dreamtonics has lagged behind in its development of English voicebanks, leaving Eleanor, Natalie, Anri and Solaria as the only feminine options for years; which no real choices for common genres like RnB, heavier EDM, a ton of vocal trends in pop, pop punk...etc...etc.... Combine that with nonsensical ads that fail to explain the product, terrible visual design for their in-house vocals, and poor SEO making their outdated SynthV R1 appear first on Google (so many potential customers have bounced off because they accidentally downloaded the wrong program) they made it easy for competition to come in and eat their lunch.

Click to expand...

This is definitely how SynthV took over Vocaloid outside of Japan. Interesting!
I think that most ACE users are (sadly) AI and Cryptobros themselves so I doubt they’ll care about the AI pictures. Most videos praising ACE (and using the program at all, unlike SynthV which western vocal synth fans actually care about) are usually “OMG AI!!!” type videos, so I don’t think that’ll be a problem for it’s userbase. Also the on the Clara Solace video they clarified that vocal synths are ethical AI. (I also originally mistook the photos for stock photos myself..)
They originally removed the AI character art due to us (vocal synth fans/fandom/enthusiasts) but I think they found a new way to keep themselves afloat—pandering to AI/cryptobros (lol)

yeah, synthv really needs to step up its game at this point. ace studio is really amping up the variation...it's definitely got a more colorful and interesting ui (synthv just sticks with the gray and green, and only added the option to color the tracks (and ONLY in the track view) later on (R1 had full color customization, so I'm not sure why R2 removed that)), and some very interesting-sounding vocals to boot, including ones that aren't usual in vocal synthesis.

This made me curious enough to see what the newest ACE vocal teaser was like. From a genre standpoint, I didn't really feel that they covered a significant amount of ground that SynthV hadn't already, but the tuning stood out as really realistic and it didn't overstay its welcome. That might address a common point of criticism from the professional sphere I see towards SynthV developers (both first- and third-party), which is that they don't have enough quality control for their demo songs and end up giving people a weak impression. On the other hand, it's hard to tell if it really is the tuning or if the natural timbre is actually better without a lot of material.

I think heavily investing in the YT tutorials community was still the most impactful marketing decision they've made so far. I'm not sure if they've ever brought SynthV out for comparison in any of those, so there's a good chance that a non-negligible amount of people literally aren't aware of the competition.

But I guess with the subscription model, ACE seems to mainly appeal to people who just need a voice for their songs and don't necessarily have the incentive to get attached to any particular vocal.

AmazingStrange39 said:

But I guess with the subscription model, ACE seems to mainly appeal to people who just need a voice for their songs and don't necessarily have the incentive to get attached to any particular vocal.

Click to expand...

It does seem to come up when those engines are being compared, the concept of owning one vocal vs. renting all of them. There's an obvious incentive to prioritize buying over renting in the vsynth community, but outside of that, it's possible that the two could coexist in different economic niches if they remain comparable otherwise. There are some who'd probably consider it a bigger inconvenience to have to pay for each individual VB.

Then again, the professional English-speaking vocal synth demographic isn't that large as it is, so it's dubious whether taking just a fraction of that would be enough to keep a company afloat.

spritze said:

It does seem to come up when those engines are being compared, the concept of owning one vocal vs. renting all of them. There's an obvious incentive to prioritize buying over renting in the vsynth community, but outside of that, it's possible that the two could coexist in different economic niches if they remain comparable otherwise. There are some who'd probably consider it a bigger inconvenience to have to pay for each individual VB.

Click to expand...

And I guess ACE's model might make it closer to "working with a singer agency but easier"; from what I've read the subscription model helps pay the devs so they'll never implement perpetual licensing (the Clara video also implies it's paying the VPs)

But it also might make it more "realistic" to the experience of working with real singers

also, with cloud services like this, vocals that are discontinued disappear entirely, which is sadder but also more realistic to real singers leaving singing (i'm pretty sure fibe..."died", basically)

Semi-related, but I've felt for a while that the sheer number of letter voices combined with the lack of notable presence they have (they really aren't marketed much) makes most of them totally forgettable and difficult to market. Professional producers might not be looking for the same fandom-based marketing that the classical "fanbase" is, but the product needs to either be included with the program or have an actual presence! There's a reason Solaria and Avanna were successful outside the fandom; they had actual marketing and something resembling a memorable concept.

DefiantKitsune said:

Semi-related, but I've felt for a while that the sheer number of letter voices combined with the lack of notable presence they have (they really aren't marketed much) makes most of them totally forgettable and difficult to market. Professional producers might not be looking for the same fandom-based marketing that the classical "fanbase" is, but the product needs to either be included with the program or have an actual presence! There's a reason Solaria and Avanna were successful outside the fandom; they had actual marketing and something resembling a memorable concept.

Click to expand...

yeah, there ARE fans that get attached to the letter vocals (and kevin sorta became a meme) but...i feel like having a mascot really does help sell the vocal. like, for me i can actually visualize the character singing, while with the letter vocals (or even v6's silhouette vocalists) it's hard to settle on one appearance for them

and there's something to be said about your favorite character singing your favorite songs, as opposed to some...nondescript disembodied (albeit nice-sounding) voice

i mean, real singers have, um...faces, so...

On that note, I feel like certain modern vocal synth designs seem to be subscribing to a sort of maximalism, with crazy bright colors and accessories, and I'm not sure I like it. Still others look too normal, like the AUDIOLOGIE characters, who look like realistic modern-day social media influencers. There are some that got good designs, but I feel like designs were more consistently appealing and not too little or too much back in Vocaloid's heyday.

(not that the audiologie vocals have bad designs! they're just a lot like irl social media influencers and i guess that's supposed to be relatable? idk i'm not an influencer myself)

I AM wondering if it's the rise of VTuber influence in the east and...influencer...influence?...in the west?