SynthV Synthesizer V Discussion Thread

[mary34]

Oh yeah! I was gonna post this yesterday but I kinda forgot...
Yeah okay, I think this is seriously making me consider getting this voicebank (but it will be probably later). She sounds so goood!

 

[blackout]

God she sounds divine in SV. I’m so happy Internet didn’t stop at Gumi and is keeping up the dual V6/SV releases

 

[morrysillusion]

wow, didnt expect the style of demos they gave her, gives a very different impression of what her voice could be used for-- the ballad/classical feeling for her voice type and vocal modes is interesting.

 

[WyndReed]

I love how she just magically poofed her jacket on for ballade and musical vocal modes.

 

[spritze]

Actually blown away by the conversion here! She works so well that I'd believe you if you told me she was recorded for SV first. Definitely get the sense that she's at her strongest when the nuances in her voice are given some room to breathe. I can't wait to hear her on some jazzy stuff now.

 

[Leon]

Gosh, she's so lovely. All of her vocal modes are so beautiful. I always love a power bank, but her ballade and dark modes are also really good. I'm so excited to see some more uses of her once her SV comes out, hopefully. I loved her design from the get go, but since V6 hasn't taken off so much, I haven't seen a ton of her. She's absolutely on my "to buy" wishlist

 

[wrong_thyme]

just gonna second the praise here, she is amazing

 

[lIlI]

Audiologie has also released a song for Halloween, and new merch with the designs:

big news

 

[lIlI]

More Koto! I hope being available on multiple engines increases her usage, she sounds great and I could see her being a hit if she finds the right producers. She has some of the popular qualities of Chis-A and Flower's voicetypes with added realism and flexibility.

 

[morrysillusion]

Finally, the synth v forum has returned!

Welcome to the Dreamtonics Users Forum

Welcome to the brand new official Dreamtonics users forum - a place to hang with the community, post about Synthesizer V Studio and Vocoflex, and share your work. Regarding our first forum, which is no longer accessible: an unfortunate incident caused most of the original forum’s data to be...

forum.dreamtonics.com

 

[BANKARA_TEEvie]

More Koto! I hope being available on multiple engines increases her usage, she sounds great and I could see her being a hit if she finds the right producers. She has some of the popular qualities of Chis-A and Flower's voicetypes with added realism and flexibility.

they certainly were brave not using, you know, the rap parameter for that one part that sure would've benefitted from it LMAO... it didn't sound bad without it, but i just think it's funny how all sv vocals can rap but it's so rarely shown off!!
god she sounds amazing though- cross-engine vocals aren't just good for people who prefer a certain engine over another, but i'm actually so curious about how people might be able to make use of BOTH of koto's voicebanks if possible - the same way people mix different vocaloid appends or mix a few of an utau's seperate VBs together to make a unique sound. i'm excited for VXB, because she'll probably be the best comparison :)

 

[Alphonse]

they certainly were brave not using, you know, the rap parameter for that one part that sure would've benefitted from it LMAO... it didn't sound bad without it, but i just think it's funny how all sv vocals can rap but it's so rarely shown off!!

They probably used the audio to notes thing instead. It's not a rap, it's just speech.

 

[BANKARA_TEEvie]

They probably used the audio to notes thing instead. It's not a rap, it's just speech.

fair point, but i'd say the rap param is still good for that kinda stuff! not so much in the sense that 'oh, it'd sound WAYYY better if they edited it to be like that', but moreso - like i said - how we don't often see demonstrations of a VB's rap functionality if it isn't inherently a part of the VB or the dataset... like, a lot of companies show off XLS when there's no extra data, so i think it's be a pretty cool thing to showcase as well, just in case... idk, somebody wants to buy koto but ONLY if she can pop off with the flow or something.... pretend i didnt say that.

 

[Twillby]

They've posted a cover of a Gumi Whisper demo to demonstrate Koto's Ballade mode (I love the way the description says, "Of course Synth V Gumi is in the harmony," it tickles my funny bone for some reason, no idea why)

 

[Alphonse]

They've posted a cover of a Gumi Whisper demo to demonstrate Koto's Ballade mode (I love the way the description says, "Of course Synth V Gumi is in the chorus," it tickles my funny bone for some reason, no idea why)

コーラス is backup/harmony vocals, not chorus. But that is funny.

I wish they'd stop using the tension and breathiness parameters... They don't sound good, lol.

 

[BANKARA_TEEvie]

I wish they'd stop using the tension and breathiness parameters... They don't sound good, lol.

OOOOOOOH! a chance for me to rant about one of my few niche issues with SV!
i think it's crazy how there are so many features which rely on variations in the vb's actual vocal data... but then things like breathiness and tension, which realistically CAN be enhanced through using the actual tension and breath data from the dataset, just aren't. obviously i don't know the technical side of things, but this is actually how diffsinger handles those parameters and, when it works well, it's MUCH better - say, extracting the actual breathy or powerful tones from each individual VM for a more realistic tension effect.
i say 'when it works well' because... the reason they may not have added it is because i know diffsinger's tension can actually also cause a lot of problems, and require a bit of trial and error for the best output - and i know that's a bit of a staple with AI in general, but probably wouldn't be such a great look for a commercial product such as SV. still, i think something along these lines would go a long way if implemented well, because i've always found the current parameters to be a bit lacking in how they clearly still appear to be simulated. of course a VM could be used to the same effect, but when used in tandem it would be much, much better.

 

[mary34]

I kinda have to dissagree on the tension and breathiness slander going on here, when applied in moderation it's quite nice.

 

[pico]

More than any parameter in a vacuum, I really think the root of the problem here is probably SynthesizerV’s extremely poor rendering capability for soft and whispery voices. Aspirations alone are generally nice, but otherwise the engine noise is incredibly noticeable in these applications.

 

[morrysillusion]

softer sounds definitely make engine noise really obvious, and its funny because i hadnt considered such negative impressions on the tension parameter because i do use it a lot but... i use it a lot to make the voices stronger, not softer, which doesnt produce the same kinda issue with engine noise. tbh i find little reason to use it to make a voice softer when a lot of synth v voices just get naturally soft at a lower range anyways and thats normally where id want it softer... but in recent years i have replaced a lot of my subtle uses of the tension parameter with tone shift instead which i like way more. obviously doesnt do the same thing but i realized that much of what i wanted sound wise was better achieved with tone shift instead

 

[lIlI]

Lots more Koto: