• 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.

SynthV Synthesizer V Discussion Thread

Bookworm2

Your friendly neighborhood Vocaloid nerd
Dec 1, 2024
92
The depths of my computer
Is SynthV 2 Cross-lang different from SynthV 1 Cross-lang for synthv1 banks? And if it is, if you use cross-lang in SynthV 2, then port it to SynthV 1, would the result be in the SynthV 1 style or SynthV 2 styl?
 
Last edited:

pico

robot enjoyer
Sep 10, 2020
590
Is SynthV 2 Cross-lang different from SynthV 1 Cross-lang for synthv1 banks? And if it is, if you use cross-lang in SynthV 2, then port it to SynthV 1, would the result be in the SynthV 1 style or SynthV 2 styl?
Is your question about moving project files from SV2 to SV1?

Project files contain data that tells the engine what notes and dynamics you want it to sing. It doesn’t contain any “singing” by itself.

The actual methodology for XLS is contained in the engine software. The “style” of the singing is predicted by the “brain” of the engine. Therefore, if you take a file from SV2 to SV1, the file is now talking to SV1’s “brain”, not SV2’s. The SV1 “brain” does not know how SV2 thinks. Therefore the style will be SV1’s.
 
  • Like
Reactions: Bookworm2

AmazingStrange39

Miku-Avanna-Gumi enthusiast
May 23, 2019
346
My SynthV 2 trial just ended so...
-Because I have Yamine Renri (SV1 standard voicebanks are ported as SV2 "Plus" voicebanks, and Renri is the only free full SynthV), I can still use the feature-limited version of SV2.
-I am not allowed most of the tuning and customization options (such as pitch bend and parameters; control points don't work on Renri anyway), but note-bending works...off and on (sometimes the autopitch fights it). Sadly I can't apply vibrato, except by painstakingly note-bending it into existence. If need be I suppose I can tune with Vocalshifter or something.
-Unlike SynthV Basic, cross-lingual is available, but only track-wide; note-level cross-lingual is grayed out as are most of the other note settings.
-It does not restrict you to a 40-note limit if you're not using a trial.
-You're only allowed 1 track; new tracks are automatically perma-muted.
-Voice to MIDI is unavailable.
-Phoneme editing is available within the language you're working in.
-You can technically "generate takes" but I don't think it does anything?
-Rap mode is available.
-Non-trial vocals can have their audio rendered.
-VST and ARA plugin versions are available.

TL;DR:
Unavailable: Parameters, pitch bend, note settings, note-level cross-lingual, phoneme timing, voice-to-midi, >1 track, working AI retakes
Available: Track-level cross-lingual, note splitting, phoneme editing, audio rendering (for non-trials), a "Generate take" option that doesn't really do anything, standalone and VST/ARA plugin, Sing and Rap mode
 
  • Like
Reactions: WyndReed
You may have heard that Cheng Xiao was used by the BBC in a Doctor Who episode! While this may seem like a deeply obscure pull, I get what the musician was cooking - Cheng Xiao's specialism in Chinese Opera gives her a singing style that sounds unusual to Western audiences, perfect for this quirkier use case.

I'm shocked that more TV projects haven't utilized how unsettling vsynths can sound to people. I've seen a lot of people online say they're unsettled by how Vocaloid sounds, so like. why not use it for a horror project, or something...
 

lIlI

Staff member
Administrator
Apr 6, 2018
1,058
The Lightning Strike
I honestly feel like there's a gap in the market for a vocal synth that's designed to sound alien - an easy-to-use vsynth that's deliberately 'off', not because of any unintended technological limitations. Producers shouldn't have to resort to Chinese opera for what is the most common creative reason an ordinary musician would reach for an AI: to get something that doesn't sound human.
 

wrong_thyme

🕯️praise gakupo🕯️
Jan 12, 2022
124
I honestly feel like there's a gap in the market for a vocal synth that's designed to sound alien - an easy-to-use vsynth that's deliberately 'off', not because of any unintended technological limitations. Producers shouldn't have to resort to Chinese opera for what is the most common creative reason an ordinary musician would reach for an AI: to get something that doesn't sound human.
i think po-uta was an attempt to appeal to that market but:
1) he's on v6
2) he doesn't actually sound that different from the regular Cute Girl vocal oops
 

lIlI

Staff member
Administrator
Apr 6, 2018
1,058
The Lightning Strike
According to a fan who emailed Dreamtonics, Anicute has ceased functioning as a storefront, and their products have been discontinued as a result.

It's very unfortunate for fans that Anicute chose not to post any form of announcement prior to this happening, although very much in line with their consistently negligent behaviour. Anicute has, throughout the entirety of their tenure with SynthV, been an atrociously run company.

Despite the endless patience and generosity of their fans, they consistently failed to deliver accurate (if any) communication, to handle their customers' money with respect or care, or to maintain any form of reliable customer service. Half the time, they could barely deliver their products.

With an increasing number of tiny, fan-run teams delivering amazing vocal synths with the utmost care and respect for their customers, Anicute is hard to excuse.

It's likely for the best that they've disappeared as a stockist (for now), but a sad day for fans of their characters, especially those who missed the chance to purchase them.


 
Last edited:

BANKARA_TEEvie

voxbox guy
Jun 24, 2021
57
...sorry to be the bearer of more bad news, but i've just learned that ST-Media may also be in some hot water for a situation similar to KAFU SV - that is, UNI's voice provider hadn't expected the result of the voicebank to be so close to her voice, so there are now talks of a lawsuit and, of course, the possibility of her voicebank being removed from sale.
i'll try and gather actual sources if possible, but sadly as i'm not using X anymore, it's annoyingly hard to do so :<
 

Leon

AKA missy20201 (Elliot)
Apr 8, 2018
1,145
Article

I respect VPs making choices about their own voices, but I call BS on not knowing how the product would sound TBH. And unauthorized AI replication?? I am so positive that AI would've been in the contract she signed, not to mention in the product name. StMedia was quite open with everything in their livestreams, they made voice demos before release, and in any case SV Uni sounds like a somewhat more realistic Vocaloid Uni imo

She's already voiced a synth before, what could you expect...? At least Kaf backed out of SV KAFU, I don't know how much I respect going through the whole process, getting royalties from the product, and then suing a small company and running to a big news group to break the news.

Idk, I guess I don't know enough to speak on the situation. But I feel bad for StMedia, who clearly had such passion for this project and were willing to produce Uni practically at a loss. My opinion at the moment is pretty uncharitable, but I know it may change if we got more information
 

Users Who Are Viewing This Thread (Users: 0, Guests: 2)