If that's the case, then why are they not releasing ONE? It's completely nonsensical to drop her Japanese singing voice and carry on with just her talk voice. There was a Chinese voice but it was just a Sinsy vocal, so I don't think they planned on using that one anyway.
I do not know how "Fumito Fumizuki literally talked to Oura at Voice Connect" can be any more "for a fact" than it already is. He also said someone else they know personally asked the same thing and got the same answer. And I already showed on their announcement page that it doesn't actually list IA English. Like, feel free to ask Techno-Speech's Twitter if you want to see if they will provide an explanation. They actually do reply in English sometimes.
I finally found the link, this was about the Chinese voice being from Sinsy.
So basically, CeVIO AI used whatever voices they could get their hands on for each language and everyone assumed that the voices they used reflected actual, future voice banks for the engine. Not just a proof of concept to use to market to companies wanting to make voice banks.
1st Place does have English support (cause of that Facecbook account they have or whatever for touring purposes), but I feel like maybe they're at the stage where they are more worried about marketing the characters/merch? I don't understand why they wanted to jump over with the talk voices to new CeVIO only, did they get cut a good deal by Techno-Speech or just the fact that it's not a secret their talk voices don't sound too good and it would help them with their YouTube channel/ONE and IA being VTubers? Or maybe they looked at the usage/sales numbers for IA English and were like "Yeah, no one is using this. Let's not pay like $60,000 to move her to CeVIO AI."
I really don't think that Techno-Speech has a big worry about marketing voices to average Chinese and English users and would rather focus on Japanese voices/market. Since they haven't really cared this entire time/could've made a couple voices other than Tsudzumi/Satou/Takahashi this entire time (they only seem like they will care if a company comes along for those languages, and they aren't even actively attempting to get companies to work with them).
(Hopefully my ramble makes sense.)
I feel like updated IA/ONE Song Voices are just a matter of time and they simply dont have the green light to confirm them yet. Remembering how ONE only first released with a Talk voice and only had her Song voice released several months after with no previous information whether it was even a thing or not (if I'm not mistaken?) so I don't think a Song voice update is completely written off yet. Talk voices are most likely their most used Cevio banks so it makes sense they'd go first I guess. I'll remain positive until their anniversary which is when these kinds of things usually get announced
It could be that they don't have the resources to update them all at once be it time or money or man power. Just look at crypton they usually can only do 3 in a generation. It definitely seems like a more involved process than hit the update button. Cevio isn't cheap that could be why they haven't release IA English because it's a higher level of risk. Really excited to hear the AI talk banks.
I just want to jump in here and say that when I contacted Techno-Speech a few months ago, they actually do a lot of the work for you (because of the machine learning being so involved). You have to send them several hours worth of singing or speaking data along with phonemes and they put it together for you. It costs a staggering amount for the talk voices (somewhere around the cost of a 2020 Honda), but it costs twice as much for song voices, and it costs extra for them to give you licenses (you pay about $1k for a certain # of them) AND it adds even more to the cost for them to create installers for you. (And no, I did not have to sign an NDA to be informed of this information, but I don't want to say what I was told flat out.)
I can't remember which post it was at, but last year or so, 1st Place had gotten help from the government in order to spread culture/encourage tourism through their IA/ONE concert activities (basically, they presented information to government in order to receive money to help their world touring activities and it was documented in an online PDF, but I don't remember where to find it again). So I'm trying to say they probably don't have tons of money to be throwing around.
I guess I could see 1st Place using money sold from their talk voices to eventually make song voices, but I can definitely understand them not wanting to pay that much all at once for two song and two talk voices given that the price nearly knocked me into the grave when I heard it. But it seems confusing why they are okay with upgrading the talk voices and not the song voices right now, I still think it could be the fact that talk voices are popular right now in the Let's Play/skit community compared to the low amount of users buying IA English and it would help out their own channel since they use the talk voices for their own skits. (ramble)
I can't remember which post it was at, but last year or so, 1st Place had gotten help from the government in order to spread culture/encourage tourism through their IA/ONE concert activities (basically, they presented information to government in order to receive money to help their world touring activities and it was documented in an online PDF, but I don't remember where to find it again). So I'm trying to say they probably don't have tons of money to be throwing around.
I guess I could see 1st Place using money sold from their talk voices to eventually make song voices, but I can definitely understand them not wanting to pay that much all at once for two song and two talk voices given that the price nearly knocked me into the grave when I heard it. But it seems confusing why they are okay with upgrading the talk voices and not the song voices right now, I still think it could be the fact that talk voices are popular right now in the Let's Play/skit community compared to the low amount of users buying IA English and it would help out their own channel since they use the talk voices for their own skits. (ramble)
I think it could maybe also be that CeVIO Talk currently seems to be considered notably inferior to it's competitor AI Talk? (I've heard a lot of complaints about current IA Talk's quality). Since talk voices are so popular, it'd make sense to have one that's more competitive to the Voiceroids.
*sigh* So apparently Satou isn't going to be sold at CeVIO AI launch, either. Can this get any more disappointing?
(I knew I was suspicious she wasn't listed on the launch page, but Fumito Fumizuki on Discord that even thought they were using Sasara for the demo at the Voice Connect event, Oura said that she wasn't going to be sold at launch. Makes me wonder if she'll be sold at all... Thanks for teasing us for 2 years, Techno-Speech. Oh, and poisoning our water supply, burning our crops and delivering a plague unto our houses.)
(I knew I was suspicious she wasn't listed on the launch page, but Fumito Fumizuki on Discord that even thought they were using Sasara for the demo at the Voice Connect event, Oura said that she wasn't going to be sold at launch. Makes me wonder if she'll be sold at all... Thanks for teasing us for 2 years, Techno-Speech. Oh, and poisoning our water supply, burning our crops and delivering a plague unto our houses.)
It was on the japanese prime minister's site https://www.kantei.go.jp/jp/singi/titeki2/tyousakai/kensho_hyoka_kikaku/2018/contents/dai2/siryou4-2.pdf
IA and ONE I can understand (or at least have come to understand heh), but no Sasara? Isn’t she 100% theirs or am I mistaken again? It feels weird for them to release the engine with, what, two third party banks and leave her out.
I don’t know what to think but I do hope we get some demos of actual banks at some pointbefore release hopefully lmao
I don’t know what to think but I do hope we get some demos of actual banks at some point
You said it... This is the worst/most confusing launch ever and it's not even out for another few weeks. People don't clearly know the difference between AI and Pro (I think a lot are assuming AI is Pro). And now we realize we don't even know what they're planning on selling other than those couple ports. I feel SO STUPID waiting TWO YEARS for Satou AI when we're not getting it (at least for the foreseeable future)... I am getting Miku NT flashbacks to when we thought we were getting Cherry Pie. Way to make potential customers feel like literal clowns.
I feel like Techno-Speech has to pull off something amazing in order to survive this terrible pre-launch or risk fading back into obscurity when we were rooting for them... Geez. Like, I am really hoping the software is cool/sounds good, I wish they would show an actual demo of the interface on YouTube for those who can't go to random Japanese events (aka pretty much everyone), but Miku NT launched without showing the interface, so I guess it's possible to be unpleasantly surprised yet again!
Also, I feel like the miscommunication was really bad on the Western side (I feel like I should've translated the article and maybe there wouldn't have been confusion, but I wasn't doing good when the announcement came and didn't have it in me to even read it properly, I just listened to what other people were saying). Meh. Dang Techno-Speech breaking my already split-in-two heart into two other twos. They probably don't even know/care about the emotional journey they took us on over the last couple days. (Yeah, I'm rambling/salty. What else is new.)
I feel like Techno-Speech has to pull off something amazing in order to survive this terrible pre-launch or risk fading back into obscurity when we were rooting for them... Geez. Like, I am really hoping the software is cool/sounds good, I wish they would show an actual demo of the interface on YouTube for those who can't go to random Japanese events (aka pretty much everyone), but Miku NT launched without showing the interface, so I guess it's possible to be unpleasantly surprised yet again!
Also, I feel like the miscommunication was really bad on the Western side (I feel like I should've translated the article and maybe there wouldn't have been confusion, but I wasn't doing good when the announcement came and didn't have it in me to even read it properly, I just listened to what other people were saying). Meh. Dang Techno-Speech breaking my already split-in-two heart into two other twos. They probably don't even know/care about the emotional journey they took us on over the last couple days. (Yeah, I'm rambling/salty. What else is new.)
Honestly I feel like there'd have still been some level of confusion even if we'd had an article translation, because they were also circulating that image of IA/ONE and Yukari/Flower all together with no indication they'd have different kind of banks (their pinned tweet isn't clear about it and doesn't link the article), and given it was announced on Twitter the picture's going to spread farther than an article. When I first looked at the picture it gave me the impression there was a chance everyone was getting both talk and song banks for AI. I might have been a tiny bit less salty if I'd read/been made aware the article at the time, but I'd still probably be pretty upset since I had fully expected IA English to be included after those demos they put out a while back (the proof of concept thing makes sense to me now after the fact but I probably would never have thought of it myself).
I'm trying to hold out hope for the release but realizing even Sasara's not on the list is making me more concerned. As much as I hesitate to use this kind of term I felt like Sasara was CeVIO's "Miku." Like, I always thought of her first when I thought of CeVIO. This release feels to me the way it might have felt if NT had started out with Teto and Ruko instead of Miku (... maybe a bit of an exaggeration but). Part of me wonders if Techno-Speech went, "Hey, Vocaloid is more popular than us, so if we release with banks for two popular-ish characters associated with Vocaloid no one will miss Sasara!" but honestly as much as I like Yukari I do have sentimental attachment to Sasara. Between that and how all the samples are with her, I fail to see any good reason to have a release without her. I'm confused as to how Yukari and Flower could be ready to go and not Sasara. Unless it really is just about expected sales/popularity.
tl;dr at this point I feel like anything goes with how this will play out and 98% of my years of optimism has been deeply shaken by these weird choices (I mean, I used to be the type to say, "Just you all wait, CeVIO will knock your socks off one day!" ... I feel a little silly admitting it now, even if I still want to believe it’ll be fine
)
I'm trying to hold out hope for the release but realizing even Sasara's not on the list is making me more concerned. As much as I hesitate to use this kind of term I felt like Sasara was CeVIO's "Miku." Like, I always thought of her first when I thought of CeVIO. This release feels to me the way it might have felt if NT had started out with Teto and Ruko instead of Miku (... maybe a bit of an exaggeration but). Part of me wonders if Techno-Speech went, "Hey, Vocaloid is more popular than us, so if we release with banks for two popular-ish characters associated with Vocaloid no one will miss Sasara!" but honestly as much as I like Yukari I do have sentimental attachment to Sasara. Between that and how all the samples are with her, I fail to see any good reason to have a release without her. I'm confused as to how Yukari and Flower could be ready to go and not Sasara. Unless it really is just about expected sales/popularity.
tl;dr at this point I feel like anything goes with how this will play out and 98% of my years of optimism has been deeply shaken by these weird choices (I mean, I used to be the type to say, "Just you all wait, CeVIO will knock your socks off one day!" ... I feel a little silly admitting it now, even if I still want to believe it’ll be fine
)
Last edited:
After read this kind of news, I really wish that Gynoid should moved to either Piapro Studio NT or Synthesizer V Studio instead of CeVIO AI.
Honestly, I kind of wish some of the banks that CeVIO has dropped so far would move to SynthV (if they don't get updated on CeVIO later), considering that SynthV's devs seem to have more of a clue about what the hell they're doing with their software.
*sigh* Western fandom was wrong yet again about CeVIO AI. Just got informed by a Japanese user that Q3 means Oct, Nov, Dec. Not July, Aug, Sep. So we look like bozos again and have no clue what's going on with CeVIO.
The project is called Hello New Dream. It has a smart phone app called Anata (You) for Dream that lets you type in words (supposed to write in your "dreams"), which get sung by Arashi (Japanese boy band). The singing is synthesized and I believe is makes them sing an alternate version to a song they are debuing called "A RA SHI (S T O R M)". The tagline for the project is "Yume dake motta tte ii desho? (It's okay if we only have our dreams, right?)" and that is actually one of the lyrics of the song.
The "project members" are 13 different companies, I think that just means they are each campaigning about the app by saying what their products can do to help you achieve your dream.
The website for Anata For Dream was down because it was too heavily accessed. I don't think I can download it anyway, because it's a Japanese app. While trying to find a clip of the singing on Twitter, I found a thread with translations:
Here are sample clips of the singing each of the Arashi members' dreams. The 1st part of the lyrics are the same, after they say
"You are my SOUL! SOUL!
Itsumo sugu soba ni aru (Always right by my side)
Yuzurenai yo Daremo jama dekinai (I won't give it up, no one can interfere)",
they read the a part of the dream (ex "Arashi no go-nin de sumitai (I want to live as one of the 5 members of Arashi)":
What the heck, it sounds really convincing???
Edit:
("I want to eat all of the negi in the world" lol)
The "project members" are 13 different companies, I think that just means they are each campaigning about the app by saying what their products can do to help you achieve your dream.
The website for Anata For Dream was down because it was too heavily accessed. I don't think I can download it anyway, because it's a Japanese app. While trying to find a clip of the singing on Twitter, I found a thread with translations:
Here are sample clips of the singing each of the Arashi members' dreams. The 1st part of the lyrics are the same, after they say
"You are my SOUL! SOUL!
Itsumo sugu soba ni aru (Always right by my side)
Yuzurenai yo Daremo jama dekinai (I won't give it up, no one can interfere)",
they read the a part of the dream (ex "Arashi no go-nin de sumitai (I want to live as one of the 5 members of Arashi)":
What the heck, it sounds really convincing???
Edit:
("I want to eat all of the negi in the world" lol)
Last edited: