What's new

SynthV Solaria is UNREAL

Word of warning - Kevin is very "square" compared to Solaria. I'm not saying its hard to get him to do a realistic performance, but his style is very straight and requires more work than I think any of the other voices, in order to make him sound relaxed.
Thank you! Just bought Solarius, Haden and SV pro.
Wrote a melody and … holy crap! Its so darn real! I used to use that old program Vocaloid. But this is AI scary good! Amazing tech. Now i will never hire a singer!
did you figure out how to get a 1 or two bar count off before recording?
It would be great if i could do that
 
Last edited:
Start your song at bar 2?
Yea.. i just figured out you can import midi files. So i can put melody track in midi file,?import into SV and then “Bounce” then drag and drop wave into DAW audio track. Most exciting tech i have ever seen in my lifetime in music. I have six songs pixar was griping about that need vocals. Now i can hire SV for the gig! Ha
Scary how real. What really blows my mind is adding my second guy voice and harmonizing. WOW two live singers. Holy crap. I stayed up all night finishing first tune.
 
if the evolution of synth voices will be the same, or faster, we can expect a huge step forward in expressiveness and other nuances... who knows what will be in 10 years!
In 10 years from now, use of music 'robots' will be a lot more common than today. AI will be used for everything: composing, playing, mixing, lyrics writing, mastering, arranging, orchestration, marketing etc – and while I understand that that kind of tech can be fascinating, I'm pretty sure the side effects of all that 'dehumanising' of the music making process will be met by many, long, public discussions.
There are strict rules re. how you can describe and promote food products. One cannot claim that there's vanilla pr lemon in drink if they are actually selling products which taste like vanilla or lemon – but are synthetic.

I can't be the only one who would not be OK with eg. 20% of all published music being made by robots? In a week there's Eurovision Song Contest 2024 in Sweden – would you be OK if the winner would be a synthetic product ?

And would you be OK if lots of the music played on radio/streaming apps was all synthesised (vocals, musicians, lyrics etc ) without you, as a listener, being told that you are listening to something in which humans hardly were involved in the music making process?

Film music somehow is a different story, because a lot of it already sounds like no humans have been involved.
 
Last edited:
Yep! That's exactly what I've been working on. A client is having me produce the music for his upcoming album, but hiring a real choir is not in the budget. SynthV to the rescue. I built these 8-piece choirs out of Solaria, Weina, Mai, Kevin, and Frimomen.

What you're hearing are all works in progress — unmixed, no lead vocals, lots of blocked-in placeholder instruments, and plenty more choir editing to take care of. In the final version, the choirs will be topped off with two real singers.

View attachment choirsamples.mp3
I swear I can hear Linda Ronstadt.
 
I'm pretty sure the side effects of all that 'dehumanising' of the music making process will be met by many, long, public discussions.
These discussion have already started ... even here at VI-C. And the industry also reacted regarding distribution ... on some platforms the distribution of pure AI songs is already forbidden. But it will be hard to find those, that were distributed nonetheless.
 
Last edited:
These discussion have already started ... even here at VI-C. And the industry also reacted regarding distribution ... on some platforms the distribution of pure AI songs is already forbidden. But it will be hard to find those, that were distributed nonetheless.
I know on YouTube Music, there doesn't seem to be much in the way of governance. I've been seeing obviously fake products on there for years. I always wonder who gets the royalties from playback, but I guess royalties are so small that no one really worries about it.
 
I know on YouTube Music, there doesn't seem to be much in the way of governance. I've been seeing obviously fake products on there for years. I always wonder who gets the royalties from playback, but I guess royalties are so small that no one really worries about it.
A different example: On Apple Music it is forbidden to distribute Atmos mixes, that are pure Stereo upmixes. And guess what ... many Atmos releases are horrible upmixes.

I'm not really astonished, because how will you avoid that. To do that every new release (every single song) must be checked and that can't be automated so far. The costs would be much too high.
 
You can make synth v sing in finnish but it takes a bit of manual work and there is a slight accent
Yeah, maybe too much manual work, considering Solaria's heavily American accent. Different thing with choirs (EW and Audiobro that I own), since they're not in the lead, but rather in the background. Much easier to blend in. But lead vocals, a lot of manual work... Maybe could be easier with some other voice than Solaria, though..?
 
Thank you! Just bought Solarius, Haden and SV pro.
Wrote a melody and … holy crap! Its so darn teal! I used to use that old program Vocaloid. But this is AI scary good! Amazing tech. Now i will never hire a singer!
Mdid you figure out how to get a 1 or two bar count off before recording?
It would be great if i could do that
Yes, Synth V voices are very good. Just wishing they’d release some classically trained voices.
 
Yes, Synth V voices are very good. Just wishing they’d release some classically trained voices.
I use Hayden! As my tenor. Snds like the tenor in Take Six. OMG when singing counter part to Solarius! Holy crap scary! You want me to post my song i wrote for my wife on mother's day with TWO singers? I did it in kitchen at 3 am this morning eating a strawberry, blueberry, raspberry fruit salad. On a lap top with touchpad. Real simple. Wrote lyrics too. Then “bounced” audio. Then drag and dropped mixed vocals on audio track in DAW, then played piano over it until i matched the feel as it was NOT to tempo. OMG truly scary! If you want i will post it? Kinda cool for first time use!
 
I use Hayden! As my tenor. Snds like the tenor in Take Six. OMG when singing counter part to Solarius! Holy crap scary! You want me to post my song i wrote for my wife on mother's day with TWO singers? I did it in kitchen at 3 am this morning eating a strawberry, blueberry, raspberry fruit salad. On a lap top with touchpad. Real simple. Wrote lyrics too. Then “bounced” audio. Then drag and dropped mixed vocals on audio track in DAW, then played piano over it until i matched the feel as it was NOT to tempo. OMG truly scary! If you want i will post it? Kinda cool for first time use!
Hayden sounds more pop/pop indie to me than classical, though I haven't gone looking for examples that might exploit his voice further. If you think you were able to push him into a sustained classical register I'd love to hear it. I've heard snatches here and there of various Synth V voices that convince me that the classical voice is not beyond the technology if only a voice was developed for it, but I haven't yet heard an example that's been able to sustain an impression of classically trained singing more than a note or two.
 
Yeah, maybe too much manual work, considering Solaria's heavily American accent. Different thing with choirs (EW and Audiobro that I own), since they're not in the lead, but rather in the background. Much easier to blend in. But lead vocals, a lot of manual work... Maybe could be easier with some other voice than Solaria, though..?
I think the best ones for it are native japanese or chinese voices since they don't add an unnecessary h everywhere
 
Hayden sounds more pop/pop indie to me than classical, though I haven't gone looking for examples that might exploit his voice further. If you think you were able to push him into a sustained classical register I'd love to hear it. I've heard snatches here and there of various Synth V voices that convince me that the classical voice is not beyond the technology if only a voice was developed for it, but I haven't yet heard an example that's been able to sustain an impression of classically trained singing more than a note or two.
The classical voice is absolutely not beyond the technology. I think the issue is probably popularity. I imagine it will be Eclipsed Sounds who look beyond the obvious high selling pop voices.
 
The classical voice is absolutely not beyond the technology. I think the issue is probably popularity. I imagine it will be Eclipsed Sounds who look beyond the obvious high selling pop voices.
Let's hope it's sooner rather than five or ten years from now...
 
That's chatgpt lyrics if I've ever heard one. :emoji_smirk:
Actually yes, except it's Gemini (Google's equivalent). English isn't my first language and I just wanted to put together something silly so I prompted it a few times, put the "best" ideas together and wrote a couple of lines of my own where I didn't like the result.

Either way, Solaria did her best with the abomination I gave her haha
 
Top Bottom