Synthesizer V (vocaloid alternative)...Christ...

[wuff_miggler]

just incredible.

apparently costs $45K to train a single vocalist.

This solaria voice is phenomenal, can go from rock like this (which i am absolutely astonished at the results of) all the way to disney.
Same users Little mermaid cover is phenomenally expressive.

There's just a little bit further to go - but - honestly if i wasnt purposefullylooking for Synth V demos- i'd have nfi this was a vocal synth. (yes - i am aware this is based on ginormous sample set for training)

[NeolithicElectrophones]

This is quite impressive. I'd like to hear it dry though. The reverb in that vid is eerie.
Definitely agree that this doesn't sound like a conventional vocal synth at all.
Pretty affordable as well. Might have to give it a try.

[wuff_miggler]

This is quite impressive. I'd like to hear it dry though. The reverb in that vid is eerie.
Definitely agree that this doesn't sound like a conventional vocal synth at all.
Pretty affordable as well. Might have to give it a try.

id be willing to bet it sounds great i'm also trying it too!
(not affiliated with the company - just genuinely impressed)
surprised there was vocaloid 5 discussion here but nothing on this synth.

[Klipspringer]

I grabbed the free version and it's pretty cool. I did a few tracks a long time ago with Vocaloid, but I've changed PC's a few times since then so my license got borked. This, however, covers it all for me. It's very fun!

[OleKim84]

Is this a real virtual vocalist ?
Is it made of samples or neural network ?
This will hit vocalists like a LinnDrum retired many drummers from pop scene in 80s;)
Joking

[wuff_miggler]

Is this a real virtual vocalist ?
Is it made of samples or neural network ?
This will hit vocalists like a LinnDrum retired many drummers from pop scene in 80s;)
Joking

i think its trained. you type words in...and it sings them
you can tune the "manner" as well...its wild.

[pricklyrobot]

Anyone else tried this yet, and have some firsthand thoughts?

I’m hoping you can tweak it to sound weird and unnatural, in new and interesting ways;)

[wuff_miggler]

Anyone else tried this yet, and have some firsthand thoughts?

I’m hoping you can tweak it to sound weird and unnatural, in new and interesting ways;)

i need to get off my ass and do so...will report back...warning: i am extremely scatterbrained and this may be a few weeks later.

[Samatha Dubs]

It's impressive. Listening to some other Solaria AI tracks on Youtube, I quickly got tired of the sound, though. The polished, generic "same-y" qualities become obvious after a while, but I can definitely see the potential in this for generating license free music. A lot of the easy listening music on generic nighttime radio channels already sounds like this. In fact, the AI might have been trained specifically for this purpose.

[wuff_miggler]

It's impressive. Listening to some other Solaria AI tracks on Youtube, I quickly got tired of the sound, though. The polished, generic "same-y" qualities become obvious after a while, but I can definitely see the potential in this for generating license free music. A lot of the easy listening music on generic nighttime radio channels already sounds like this. In fact, the AI might have been trained specifically for this purpose.

i think its use for its intended purpose might be the most boring application.
where i see it shine is extreme "mis-use" ie....rapid LFO modulation of wierd things like vocal gravel, falsetto/head voice transition, and so on.

[fitzgreyve]

This is the best example I've come across, really shows off what can be done with skilfull programming:



The free "lite" versions of the voicebanks offer a lot less flexibility than the full versions. If relying on these "free" lite ones, there are only 4 english speaking female and no male as yet.

[Samatha Dubs]

where i see it shine is extreme "mis-use" ie....rapid LFO modulation of wierd things like vocal gravel, falsetto/head voice transition, and so on.

If you decide to that, I'd love to hear what you come up with

[wuff_miggler]

where i see it shine is extreme "mis-use" ie....rapid LFO modulation of wierd things like vocal gravel, falsetto/head voice transition, and so on.

If you decide to that, I'd love to hear what you come up with

how the fuck on their site - do they put "out of stock" its a fucking download....fucks sake some companies...

[wuff_miggler]

where i see it shine is extreme "mis-use" ie....rapid LFO modulation of wierd things like vocal gravel, falsetto/head voice transition, and so on.

If you decide to that, I'd love to hear what you come up with

i have a secondsound unisyn on the way in a few days - i just realised i could literally "sing in" the midi myself....repitch it in dreamtonics - then play with the words....i THINK this will be insanely natural - will post asap.

[Samatha Dubs]

how the fuck on their site - do they put "out of stock" its a fucking download....fucks sake some companies...

Japanese English can be challenging

It probably just means they decided to limit the number of free downloads.

The UniSyn looks like an interesting device. It actually sounds like a sensible use for it. And if you wind up creating a generic pop song, then that's OK too. I'll look forward to it:)

[wuff_miggler]

^ the midi performance is MUCH different to the CV outs....the CV outs are 1:1 following my voice - really fucking fast -it's astonishing actually...
midi not so sure - i had a suggestion to the dev, waiting to hear back from them