[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/vt/ - Virtual Youtubers

Search:


View post   

>> No.78188353 [View]
File: 174 KB, 1330x1299, 1717789847393950.jpg [View same] [iqdb] [saucenao] [google]
78188353

>>78188003
I learned nipponese to play smut VNs.

>> No.77596682 [View]
File: 174 KB, 1330x1299, かわいいですね.jpg [View same] [iqdb] [saucenao] [google]
77596682

>>77596609

>> No.66019703 [View]
File: 174 KB, 1330x1299, かわいいですね.jpg [View same] [iqdb] [saucenao] [google]
66019703

>>66019184
I don't really care if it was Ved or some other person that makes the a.i. I just ask for it to be palatable and not robotic.

I.e. I cannot stand neuro v1's songs due to the robotic voices you get from it. But the base Hiyori model, the speech flow from azure's 20% pitch 5% rate, and whatever prompt vedal has for the old neo-gpt LLM made it palatable enough for me to watch and feel entertained enough to continue it.

I'll even go against the grain and say I wasn't the biggest fan of hearing evil's (v2) voice at first. Only really starting to like it due to the high variety of pitches and shit it can do. But if evil's voice (could just be me now, but, current evil's voice sounds more fine tuned) was the first thing I heard, I probably wouldn't have watched Neuro.

>> No.65618896 [View]
File: 174 KB, 1330x1299, かわいいですね.jpg [View same] [iqdb] [saucenao] [google]
65618896

>>65618564

Navigation
View posts[+24][+48][+96]