WebKit

Discover the ImDontai RVC v2 model, an advanced AI voice model trained over 330 epochs for exceptional accuracy. Originally intended for music generation, this versatile tool also excels at creating AI covers. Trained on a 48k sample rate with a unique 12:10 dataset derived from one of ImDontai's videos, it boasts remarkable precision thanks to TensorBoard monitoring. Unleash your creativity and generate text-to-speech or AI covers effortlessly using our free AI tools.

I used tensorboard for the first and thought the model overtrained. It didn't but I dont feel like redoing the model. RMVPE 48k Sample Rate 12:10 Dataset Dataset grabbed from one of his videos

ImDontai

ImDontai | RVC v2 | 330 Epochs

resizable

<@871469407057096705> Isn't the instrumental copyright? You can't use an instrumental of a copyrighted song.

notthecalmperson

i did see this instrumental used in roblox game, but im not 100% sure

I'd just check and it is copyright, you going to have to remove the instrumental. (Or if you think you can make a cover of it, you can try to do that.)

reuploading bc one of the samples didnt abide by the rules. rmvpe 48k sample rate 3:13 dataset dataset grabbed from episodes of the mk series (episodes 1, 2, 4, 5, and 6)

Baraka (DashieXP MK Series)

Baraka (DashieXP MK Series) | RVC v2 | 400 Epochs

Another remake of a model because why not. This one came out really good. RMVPE 16,650 Steps 40k Sample Rate 15:02 Dataset Dataset ripped from one of his youtube videos

Daz Black/Daz Games

Daz Black/Daz Games | RVC v2 | 450 Epochs

I surprised this didn't overtrain lol. Would've trained it for longer but I was scared that I would lose connection runtime. This model came out decent but at the same time when it pronounces the letter "S" it sounds kinda weird, so Im sorry for that. RMVPE 48k Sample Rate 28:07 Dataset Dataset is raw audio of some of his quotes from Bully SE

Jimmy Hopkins (Bully SE) | 22

Jimmy Hopkins (Bully SE) | RVC v2 | 300 Epochs | 22,800 Steps

I forgot the amount of steps this took lmao. doesnt matter anyway RMVPE 48k Sample Rate 00:31 Dataset Dataset taken from a YouTube video

Singing Persian Turtle

Singing Persian Turtle | RVC v2 | 750 Epochs

Remake of my first ever published model I realized it was starting to overtrain at 400 epochs RMVPE 10,530 Steps 10:07 Dataset Dataset grabbed from two of his vids Link

ImDontai

Description

Samples

Saiba Momoi (Blue Archive)

Adel imam (Egyptian Actor) عادل امام (ممثل مصري)

JP Furina / Focalors CV: Inori Minase (Genshin Impact)

Ariana Grande AI

Hu Tao JP (Genshin Impact)

JENNIE of BLACKPINK [Strong Ver.]

Saiba Momoi (Blue Archive) (VA: Tokui Sora)

Hatsune Miku

SpongeBob SquarePants (Talking And Singing)

Takanashi Hoshino (from Blue Archive)

Satoru Gojo (JJK) [VA Yuichi Nakamura]

ENHYPEN Heeseung

Sunaokami Shiroko (Blue Archive)

Goku (Mario Castañeda) (Latin American Spanish) (UPDATE)

Villager (Minecraft)

Mortis [Brawl stars]

Jungkook (BTS)

Melodie - (Brawl Stars) - (English & Korean Voices)

Tendou Arisu (Blue Archive)

Kanye West