SynthV Synthesizer V Studio 2

[morrysillusion]

im confused how the download is being accessed? are people activating their serial code and getting it? because i have the code for the upgrade version and all i get are errors when trying to activate it. plus the fact the new my.dreamtonics.com doesnt show any of my products... do you have to actually add all of them again or what?? additionally if that is the case... is there some easy way to find all your serials through whats active on your pc? i have loads of my emails but, i cant find the code for my actual copy of the SV1 editor. (or if i am, theyre not at all in the format that is current, from some really old volor emails)

 

[BiscuitsandTriscuits]

You have to add every dreamtonics product you purchased from them into your account with the activation code then you will apply the upgrade to the pro editor to upgrade it (same with voicebanks if you buy the upgrade)
I searched dreamtonics in my email to find my codes and it went well just a warning if you have a 3rd party vb you have to upgrade through other means I haven’t figured out yet e.e

 

[morrysillusion]

I have successfully activated and upgraded all my voices through the new site but, i own Synthesizer V Pro 1 and did just manage to find the correct code (i think. i bought it through anicute ages ago when it first released). but still seem to be getting errors :/ guess i will email support

somehow it seems to be saying its already "bound" to an account, and im not sure how that is or what account it would be. considering i didnt buy it off the dreamtonics store, it wouldnt have been activated through any account on there, and you couldnt do this sort of activation before...

 

[Leon]

You never traded codes with people online, did you? Or bought a secondhand code off someone? They opened the my.dreamtonics thing a while back, so people were rushing to activate register their serials to prevent others from doing it...

Also throwing it here instead of just leaving it in my status, but IDK if it's just me, or an issue on DT's side, or something with Beijing Photek, but Stardust fails to install in SV2 every time I try. Everyone else installed fine.

 

[lIlI]

I have successfully activated and upgraded all my voices through the new site but, i own Synthesizer V Pro 1 and did just manage to find the correct code (i think. i bought it through anicute ages ago when it first released). but still seem to be getting errors :/ guess i will email support

somehow it seems to be saying its already "bound" to an account, and im not sure how that is or what account it would be. considering i didnt buy it off the dreamtonics store, it wouldnt have been activated through any account on there, and you couldnt do this sort of activation before...

I know this can happen if you ever traded codes, and someone else activated one - but I've also heard of it happening through a glitch.

Hearing through the grapevine that Mai has 6 vocal modes now, that makes her a more competitive choice as a free voice. I'm looking forward to hearing them!

 

[morrysillusion]

I know this can happen if you ever traded codes, and someone else activated one - but I've also heard of it happening through a glitch.

i have never traded codes ever :/ i got Synth V before it was Pro editor from Anicute and Anicute had offered their own sorta upgrade discount when Pro was coming out-- highly doubt there is a difference there... but i do notice how they called it "VIP", maybe they did something different that makes it not compatible?? idk. hope i get a response soon but tbh debating if it might be faster to just refund the upgrade and buy it full lol... the cost isnt a huge sink for me so

 

[Alphonse]

I know this can happen if you ever traded codes, and someone else activated one - but I've also heard of it happening through a glitch.

Hearing through the grapevine that Mai has 6 vocal modes now, that makes her a more competitive choice as a free voice. I'm looking forward to hearing them!

I was looking for Mai's new vocal modes earlier but didn't see them on the main website or the store. In fact, the main website only had a sample for one of her original vocal modes, Soft. Lol.

 

[amemelia]

mai's new vocal modes are breathy, downer, emotional, soft, rap, and sweet. not what i expected, excited to try them out!!

also if you happen to have a bajillion dreamtonics banks like me and bought the 5+ vb upgrade code, you can *re-enter* it to activate mai 2!

 

[Alphonse]

also if you happen to have a bajillion dreamtonics banks like me and bought the 5+ vb upgrade code, you can *re-enter* it to activate mai 2!

What do you mean? Are you sure this is separate from upgrading the engine and getting one of the free voicebanks?

 

[Leon]

What do you mean? Are you sure this is separate from upgrading the engine and getting one of the free voicebanks?

It's for people who didn't choose Mai as their free SVP2 bank. She wasn't added to our my.dreamtonics pages before, but now you can purchase her upgrade. And if you already did the one with the highest number (was it 5?) then you can put it in again to get her SVP2 too. I'm glad! I was actually thinking about it the other day, how if you already upgraded your alphabet soups, Mai wasn't an option at the time

 

[amemelia]

yes, what leon said-- sorry i wasn't very clear! i grabbed liam as my free voicebank and was relieved i didn't need to spend another $20 on mai now to decide whether that $20 is worth it for jun nocturne in the hopes that it fixes his annoying autopitch...

 

[parallax_fifths]

I can't find a full screen mode... All my other music software has one, but not SVS2? (Or SVS1 for that matter.)

 

[pico]

Will need to dig in further to create solid thoughts on how I feel about the audible characteristics of this new engine and the nitty gritties of the features, but I'm actually fucking stoked right now. Was not expecting that, at all. The posts I saw down the grapevine from friends made me think the engine might fall apart at the seams when I launched it. (It still has the opportunity to, but point being, hands-on first impressions have been great for me.)

I will create a full track and tune it over the weekend with the intent of discovering as much jank as possible, but so far, the engine has been behaving nicely. Dreamtonics' new account system, on the other hand, will redirect you to empty API endpoints. Putting that aside for now.

• Performance so shocking I actually almost choked: I'm on a laptop spec'd for low-mid end scientific simulation and modelling tasks. Every single piece of voice synthesis software will choke and buffer on rendering. SV1 was among the best in this area, but the render times did still interfere with workflow. When I say I am experiencing ZERO render lead times on this engine I am not joking. This shit is insane. The re-processing when you adjust a global parameter still occurs but it happens so fast that by the time I move the playhead there it's done.
  • I expect this may not be the case to such a dramatic extent with longer tracks, but man. Makes me want every engine to have proper memory handling and multithreading so bad.
• Mega gripe addressed: although not perfect, the intonation and feel of soft/whispery vocals appears to have improved massively. The presentation of these voice types and expressions was so grating to me I would actively avoid using it in SV1. It made me not like Chifuyu at all, as well as Moka's soft expressions. This was tragic as cute girls are my everything. Fingers crossed that it remains improved for 3rd party vocals with SV2 versions.
• Vibrato adjustment is good now: I absolutely did not like the vibrato adjustment in SV1, which poorly addressed the global vibrato modulation across groups of notes. Beyond that, adding vibrato to a note procedurally added modulation to the note end in a way that felt completely unhelpful to me. Now, turning vibrato off actually turns it off and turning vibrato up actually appears to scale the model's vibrato prediction instead of forcing additional modulation on top. Win.
• Manual tuning: I actually think the new setup is a far better solution for manual tuning and intervention than the old one... hear me out.
  • To get the big question out of the way, Dreamtonics' assertion at NAMM that the same environment can be created by setting the notes to "rigid" and vibrato to '0' achieves the same effect as manual mode seems completely true. The basic modulations, that were present in SV1, will still be there-- so the gentle lead in/off before pitch transitions that was present there is intact. For those worried about that aspect, I can say that at least with Mai 2, this absolutely works.
  • I don't know if anyone else had this perception about SV1 but I felt that switching into and out of manual and auto modes on a note or group of notes was an absolute nightmare. Converting an auto note to a manual note created an absolute mess of pitch modulation. Deleting the modulation wouldn't create something consistent. Going the opposite way felt like it created a full range of awkward behaviors. Now, everything is integrated, and this phenomenon seems to have been accounted for in the design of the software itself.
    • I know some folks will be against the absence of a strict manual mode on principle, but for me this is the best case outcome because it seems it reduces awkward interactions where the software feels like it doesn't want you to interfere/do manual things. I think AI intervention to some extent has always been non-negotiable in the design of SV Studio. These folks are developing a genAI product and there is no escaping that. This is why I remain very excited about OpenUtau and NT2, which provide concatenative-first environments for that classic style and workflow.
• Parameter 'layers': Parameterization like layers in a bitmap art/photography software. Cool and UX wise feels better than old system. Haven't played with the parameters themselves in depth yet.
• Control point tuning on timeline: I really like it
• Question mark: My biggest gripe with SV1 remains the homogenization of its vocals. So far I don't feel like 'raw' is doing much to Mai, but it could have to do with Mai being an in-house vocal the engine is being developed around/for.
  • I will wait to see what happens to SV2 third party vocals to make a judgement call here. My fingers remain crossed but I'm admittedly kind of not optimistic here. It is in Dreamtonics' business interest for everything outputted by the engine to be "good" by a general audience metric. I believe that is why this symptom exists in the first place. We'll see

Again withholding judgements on vocal output for now, too early for me (aside from the soft vocal thing which stood out immediately). But from a UI/UX perspective I genuinely love this. Shoutout to the UX people (or person-- whoever you are) and testers. It appears to have paid off for real.

Laptop specs for those curious

processor: i7-1050H
16GB ram
integrated 'optimus' graphics, which I hate.

 

[Leon]

I need to play with it more, but IDK, mine has been rendering kind of slowly actually. To be fair I hit it with a doozy of a 7 minute song, but all I had was the music and like less than half the lyric notes input. It would load for a solid 30-60s with just some smaller tweaks, and before I gave up for the night I sat for over 3min before deciding to try another song and having the editor stop responding.

It's better with shorter songs, at least for now and in my experience. FWIW, I'm not using a bad computer. 64GB RAM, Ryzen 7700 8 core processor, Radeon RX 7800 GPU. Chrome always has too many tabs open, but I was sitting around 64% memory usage with SVP2 doing 1GB more than Chrome when it died on me. RIP

I'm enjoying it otherwise. I want to play with it more before making any more technical comments.

 

[Heruru Meruru]

After playing around a bit the output is more "human sounding" for all that entails, both good and bad. Solaria's vocal modes also sound more distinct. V1s also sound smoother. I really love playing around with the expression settings (and they work with V1 banks!) but it is true the cross-lang accents are a lot thicker, especially Yuma who is barely intelligible in English. I hope they manage to fix it.

Rendering is slower for me but my PC is kind of potato-y. The knobs are a bit confusing to control, you have to move your mouse up and down rather than dragging... I'd rather just have a gauge like v1.

Can't wait for Hxvoc and Galenaia to go live. I forgot they were releasing on the 27th and thought they were going to be launch voices.

 

[Alphonse]

We kinda knew this from the new Eclipsed Sounds vocals' demos but I think it's worth noting that vocal modes seem to interact differently now. It sounds like they are more additive now, compared to before which was more averaging between the two modes than anything. Like, I'm trying Mai's Downer and Sweet modes and if I put them both at max, I get the same level of softness of Downer at max and roughly the same level of cuteness as Sweet at max. Same with Emotional and Powerful: if I put both at max, I get 100% of the fullness of Emotional and 100% of the power of Powerful. And like Eclipsed Sounds mentioned, if you put two soft modes at max (like Mai's Breathy and Downer modes), you get something softer than either of them alone!

Before, having one vocal mode at max and putting another at max would lessen the effects of either of them, so I really like this change.

 

[Rylitah]

If you were having trouble registering a vocal to your Dreamtonics account, the issue was on Dreamtonics' end - AHS made an announcement about it, then made another one just now saying the issue has been fixed.

https://x.com/ahsoft/status/1902962033647333743

(As someone who had this issue when registering Teto to my account yesterday, the exact message was 'Error 400: Invalid Activation Code' - after the fix, I was able to register her just fine. If you had the same issue as me, it should work now!)

 

[Palamon]

Can't believe they released it in a super buggy state like this unbelievable.

 

[Blue Of Mind]

Can't believe they released it in a super buggy state like this unbelievable.

Most new software is really buggy on release - that's why so many people don't preorder or buy games on launch like they used to, for example.

EDIT: Remake of Mai's original demo with SV2.

 

[lIlI]

Saki Std and Renri now have plus voicebanks up, meaning they can sing in all languages now! Quadimension plus voicebanks are also available, but they must be purchased. *edit* The other Volor banks too!