testing my new voicebank, again.
Vložit
- čas přidán 13. 01. 2024
- Hello there. It's me, Kureon-P! And I am back with another video after being inactive for a considerable amount of time. For a little bit of context, I was testing the other kits.ai voicebank I created that had been laying dormant for a few months along with the younger-sounding version of ALPHONSE it's about time to tell you that ALPHONSE actually had two versions. It's just that I tend to use the younger-sounding ai model more than the other one, which in turn made me forget to showcase ALPHONESE's other model which is called "DEBUT" since it corresponds to his very first model. Speaking of which, this is also the first time that I used a different input. Usually, my inputs come from the recordings of my voice, acapella vocals from existing artists and/or celebrities and utau... Now that I used SynthV as my reference vocal for the first time, my first impression was: "This sounds ok, actually this is much much better that the other sources where I get my reference vocals from.". Anyways, this is just a short test.
Setting that aside, a little announcement about an utau cover with the incoming release of my very first UST, yes, it is really coming but you have to be patient because again, I am quite slow at stuff. But I will make sure to put it up on CZcams for you guys to see my upcoming work regarding that.
Anyways, I hope you'll like this video :DDD!
|/|/|/|/|/|/|/|/|/|/ :D CREDITS :D |/|/|/|/|/|/|/|/|/|/
Song: Perfect Night by LE SSERAFIM
Reference Vocal: SynthV - SOLARIA
SVP & Tuning: PixPrucer
Cover & Mixing: me
Voicebanks used: ALPHONSE_CHILD & ALPHONSE_DEBUT (by me)
Video: me
Engine used: kits.ai
Dataset: 10 minutes
|/|/|/|/|/|/|/|/|/|/ :D LINKS :D |/|/|/|/|/|/|/|/|/|/
My website: https//sites.google.com/view/kureon-p
Reddit: https//www.reddit.com/user/KureonUTAU
My Discord server: / discord
Tumblr blog: tumblr.com/blog/crayonshincha...
Try out ALPHONSE's models here:
app.kits.ai/convert/shared/al...
app.kits.ai/convert/shared/be...
Oh, and here is his wiki entry btw:
vocadb.net/Ar/130399
I do not intend copyright infringement! The medias used in this cover is credited to the original owner! - Zábava
![reality vs expectation but i applied modulation to it [utau shitpost]](/img/n.gif)
If you are going to use ALPHONSE in your covers, then reply to this comment with a link to your cover using ALPHONSE so I can watch it too!!
Also, I believe I already said this countless times already, but for those new viewers watching my video, I am not using Synth V in this cover. I used kits.ai, screen recorded the svp project in Synth V and timed it to the vocals to give the impression that it was being played in Synth V (Just like those show tuning covers in UTAU and VOCALOID, if you know what I mean by that.)
wtf sounds sick, I didn't expect that from a 320 views video
This was just to test my own AI voicebank a little bit and I am glad you liked how they both sound 😭
I can still hear Solaria-
Let me, once again, take the honor of being the first commentor in this video (Don't mind this HAHAHAHAHA)
fine ill go play overwatch again
LMAO
Is it possible make your own voicebank on Synth-v?
No.
Dreamtonics never supported the creation of custom voicebanks for SynthV in the first place…
No, and if you ever see "synth v port" its fake.
@@SucoTeiKun_pohtath exactly!
And I'm sure they just took voice samples from a utau or a vocaloid voicebank. Which is not a good thing to do
How'd you get your voice high pitched?
I just used sample of my voice from when I was 11-years-old 😆
Along with that, I recorded an older AI voicebank which uses my currently deep voice!
Please upload make it a full cover 😭🙏
Unfortunately, the full SVP I downloaded was cut off (I don’t why but I think it’s due to the fact that I am just using the Basic Version of SynthV). So, I can’t make a full cover…
that's okay we appreciate it! Thanks
Justin Bieber when he was younger omg
basically this voicebank is a justin bieber dupe 😭
Hmmmm.....
what do you think :3?
@@UtaKure the English is good
Though I feel like something like an Arpasing utau would have made better quality.
Don't get me wrong, this is perfectly fine
And I can't really speak for you since maybe it is easy
But it does look a bit hard to use
@@ascensiongacha1308 It's kinda easy for me since I just morphed the reference vocal into the AI model (just like Diff-SVC, so-vits-svc and RVC.). It did not undergo a much more complex process of training just like DiffSinger and/or NNSVS.
Probably you never asked this question but for the video display, I just screen recorded a SynthV project that is playing.
I can still hear Solaria-
It’s because I really used SOLARIA as a vocal reference for morphing into this model ;)