We're currently having issues with our e-mail system. Anything requiring e-mail validation (2FA, forgotten passwords, etc.) requires to be changed manually at the moment. Please reach out via the Contact Us form if you require any assistance.
I've had a pretty iffy experience with the software as of now. I didn't have much time to fully delve into it since I had to head to sleep then later school however. My only worries are that my laptop can't handle the software, which doesn't make any sense since it handled svs1 just fine. It's a brand new laptop that I got for Christmas, so I don't understand why the newer software is running worse besides possibly launch bugs.
Using vocals in the software is amazing! I spent a while starting tuning with UNI and it was quite fun! I just worry a little for SOLARIA, since after plugging her into some of her demos (Dawn specifically), her voice got kind of weird on the higher range.... I don't know if this is because I used the edited version or what. I'm planning on really delving into SOLARIA 2 tonight, and will probably update on my opinions later.
Someone mentioned her a bit ago in a friend chat and I realized I had never gotten a Renri when I filled out the google form like 2 years ago. I logged in with a different acct and filled it out again, got her in like 10min. I'm surprised they're still doing that, but I'm glad! If anyone else wants her, you go to anicute.com and scroll down till you see Renri, then click that and fill out the form.
@lIlI do you happen to have links to where the Quad plus banks are meant to be purchased? :0
Someone mentioned her a bit ago in a friend chat and I realized I had never gotten a Renri when I filled out the google form like 2 years ago. I logged in with a different acct and filled it out again, got her in like 10min. I'm surprised they're still doing that, but I'm glad! If anyone else wants her, you go to anicute.com and scroll down till you see Renri, then click that and fill out the form.
@lIlI do you happen to have links to where the Quad plus banks are meant to be purchased? :0
Unfortunately it's through their Taobao again - with the language barrier I'm not 100% sure if they're up for purchase yet; it will need to be through a proxy most likely.
Sigh. Taobao is so frustrating because you can't even look at product info without logging in, and many proxies seem to have rules about digital purchases. But I'll try to figure it out, when the time comes. Thank you!
editing to add: When I got up today SV2 had an update to run, and the Starry Court II were available. Nyl is behaving how Stardust was for me yesterday and fails to install, ditto with 2 friends who also have their upgrade. Stardust no longer even has an option to try to install, so maybe they're fixing her. The file for that longer song I tried to do last night that kept freezing and eventually crashed is working fine for me for now too
lol for AiKO's Plus update, they accidentally swapped her small icon and her full body art. so it uses the full body art as her icon and her small icon as the piano roll background
Unfortunately it's through their Taobao again - with the language barrier I'm not 100% sure if they're up for purchase yet; it will need to be through a proxy most likely.
Looked through the shop extensively, doesn't seem like they are up yet. I will immediately get Cangqiong's update, I've been waiting for this for YEARS
edit: I think they are up now! Here's the link 商品详情
I am a bit confused on which option I have to choose though, there is SV2 Cangqiong but also just Plus update as it's own seperate thing? help
Looked through the shop extensively, doesn't seem like they are up yet. I will immediately get Cangqiong's update, I've been waiting for this for YEARS
edit: I think they are up now! Here's the link 商品详情
I am a bit confused on which option I have to choose though, there is SV2 Cangqiong but also just Plus update as it's own seperate thing? help
I still can't see it (OTL), but I would hazard the guess that the Plus update is for users who already own the SV1 standard banks, and you'd get SV2 if you don't own her previous version
It took about 10 minutes for mine to come through, but after searching both of the emails I would've used, I guess the first time I filled it out, it just never came. A quick look at reddit threads of past shows a few other users with the same issue. If you don't get it within a few days, I'd just fill it out again with a different email!
ES made a video about the workflow of SV2, which is great as i often find their videos to be far more informative, especially for the crowd whos very familiar with vocsynth
i have now been able to use SV2 and they addressed some of the things i noticed-- mainly the parameter difference for tension. by god, do some of my old files using high tension sound awful lol. it kind of "distorts" the vocals, and raises the volume a lot now (to where some of my audio was peaking). it does sound good when used correctly and at times the distortion almost gives a "growl" effect, but definitely a reminder that you cant exactly plug and play old files...
I'm near tears now, friends. SynthV2 still refuses to connect to my Dreamtonics account, and I'm unable to use any voicebanks I registered . Does anyone have the same issue, too? I contacted customer support today, tho. Reinstalling the program or changing the Internet browser did nothing. It can't be an issue with the Internet connection because Piapro NT2 and Voisona work just fine, and they have the same authentication process as SynthV2.
I basically wasted 44€ for nothing, and I'd take a buggy mess as long as I could create eldritch horror sounds with Tsuina-chan...
@Leon 7 hours later, nyl still won't install for me! my monarch :(...
anyway, being totally honest, i'm much more pleasantly surprised with the release... i was expecting it to be *bad* bad with how some people were talking about it
most shockingly, i actually really love the new tuning system... i can draw over everything as i please, or just make a small mark with the pencil tool where i want to put a pitchbend and allow the system to figure out where to make it flow best!
i know everything's a bit more automated, but the workflow is certainly made up for in other areas - it at least feels less like an attempt to pander to ai bros and more like they're trying to integrate SV more casually into the *professional music* space... maybe.
the engine that the voices run on is really different, i can hear it, but it allows for a lot of human (throaty ?) inflections in the voice that make them a lot more unique, kinda like voisona... and in turn, a lot of the more automated features exist to best bring out the unique features of the voice provider, and everything else at the user's disposal is to add their changes, but being able to overwrite what's already given to you. i don't think this is a bad thing!
at the very least, i keep thinking about the fact that kanru hua is also at the lead of AHS these days, which almost feels like his way of being able to keep the synthesis stuff he does in the vsynth space.
TL;DR I really like the update and think it adds a ton of new character to the vocals!
Honestly I LOVE the way that the engine sounds! The only problems and main iffys I've had were on the dreamtonics voicebanks oddly enough. It seems like the patch sent out this morning fixed every problem I had! I've also figured out you can get the sort of "blank slate" that people were missing with manual mode by changing the expression all the way to rigid and setting the vibrato to 0. I personally don't pitch-bend, so the ai tuning doesn't bother me, and even adds more flair in my opinion. I love how much more control you get over the vocal modes and how they do so much more than before! It's also a pleasant surprise that they properly layer instead of "nullifying" themselves! This is especially fun for messing around with UNI!! I haven't tested out the AI retakes yet, but I'm sure that is real helpful too. I also like the new UI! It's not my favorite, dont get me wrong, but it gives the software the "modern" feel that dreamtonics has always sort of been going for. The only issue I have is the contextual tips sometimes still showing up even though I have it unchecked, namely when I import someone else's SVP that used a different vocal.
ok i have been fiddling with it a bit now in the last day or so, and my impressions are good. i have been able to download and update all my banks, except for NYL, which i know others are having issues with.
i mentioned it in a previous post but throwing some old SVPs in, i did need to adjust to figure out how to make them sound good, because otherwise some were uh, not good lol. but after tweaking old files, they really sound great. still getting a feel for the extra vocal mode knobs, and how they work best, but its very cool to mix things up with them. i have tried mouth opening and boy its weird but cool. vocaloid was the only program to have anything close to that and this is far more advanced technically, and its very neat. i will have to make sure i remember its there as the moments i tried it, it can make a huge difference.
the UI is very nice. i also am pleasantly surprised i enjoy the changes they have made to pitch editing etc. my method of tuning often means turning on "manual" for all my notes, where it then converts all the automated pitchbends to nodes. and then i edit/remove those as i wish. knowing they changed it quite a bit, i wasnt sure. but its really nice just changing a slider (all the way to rigid) to make things basic, and be able to edit the pitch bends as i wish, with no real harm in turning that slider back to something else.
so far i have been trying Felicia 2, Solaria, and Liam (completely forgot about Liam until i got to pick a voice). They all sound fantastic in their updates and i think the engine currently sounds pretty good. i can tell its different, but i dont feel like its really ruined any of the voices i have used.
now, something else. i know people have been talking about japanese -> english cross language and how accents seem worse/stronger. Yuma is one of my faves and his accent already is very unique and was always a bit harder to understand until later updates imo. so i decided to do a render to showcase his english from his first ever release (v100), his final sv1 release (v107), and his first sv2 release (v200): listen here
(all samples use automated tuning respective to those versions, 20% tension, low power vocalmode, slightly lowered tone shift)
to give my opinion, its definitely the worst with him imo. it doesnt sound too bad to me with some others like feng yi or yi xi (but you can tell it changed). i think yuma's tone and quality is really good, i love it actually. but the accent is very much not good for his english.
i plan to send a feedback email about this and i highly suggest you do too-- if dreamtonics wants to keep appealing to the english industry, and we find it important that their faithfulness to the voice doesnt actual ruin the cross language, its worth pointing it out. id love to see something in the middle enough that the AI isnt taking away their key features sure, but not to where they just cant be understood easily. (i think us with the ear for these voices probably understand them even better than the casual listener without realizing either)
also: doesnt seem like i can select any of the older versions of SV1 voices in SV2? like i can only pick Yuma 107, nothing else. doesnt look like there is a solid way to install them to SV2 unless i am mistaken so, guess i will use SV1 for any older versions...
the xls is a mixed bag atm... i'm wondering how much of it was a conscious decision, and how much of it was just because of the new engine/retrains (edit: w/ the nature of how the engine's output feels more like a natural recording of the voice, a la the actual dataset, i'm wondering if there's less ... 'morphing' for xls and more reliance on the vb's own data). liam sounds surprisingly nice in korean, mai's english is way cuter too - i think we'll see them responding to feedback pretty quickly tbh
By the way, in SynthV 2 they changed the location of the folder containing voicebanks, dictionaries etc. So to install custom dictionaries for instance, it would be at the location "%appdata%\Dreamtonics\Synthesizer V Studio 2" (copy paste this as the windows file explorer URL)
Oh BTW I was able to install Nyl II late last night
I haven't had a chance to check on Stardust this morning. As of last night she still wasn't available to even try. If she's still not available: Has Quad or DT said anything, does anyone know?
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
