WebKit

 Introducing our latest advanced AI voice model, "Jon Bellion (Mixtape Era) (RVC v2) 2

Trained by me, Dataset size: 13 minutes of vocals from Translations Through Speakers, The Separation, and The Definition Model

Jon Bellion (Mixtape Era)

Jon Bellion (Mixtape Era) (RVC v2) 250 epochs (mangio-crepe)

joeymag49

(Reupload) Trained by yours truly using 15.5 minutes of in-game voicelines Huggingface: Alt:

Arataki Itto (Genshin Impact) (VA: Max Mittelman)

Arataki Itto (Genshin Impact) (VA: Max Mittelman) (rmvpe, 300 epochs)

(Reupload) Trained by yours truly Please tag me if you use this for a cover >1 minute Sample rate: 40hz Training length: 200 epochs Extraction Method: rmvpe Huggingface: *Alternate * NOTES: I removed the reverb from the voice lines when training the dataset so the voice it infers will be pretty clean. To replicate the in-game sound, use a DAW to add reverb/echo to your taste. This model works best on voices with a breathy quality.

Rhodeia of Loch (Genshin Impact)

Rhodeia of Loch (Genshin Impact) (rmvpe, 200 epochs)

(reupload) Trained by yours truly Dataset 5 minutes (Cleanly iosolated lead vocals from "This is Me""Our Time is Here" and "La La Land") Extraction Method: rmvpe Sample Rate: 48hz Training Length: 460 epochs Huggingface: *Alt. *

Demi Lovato (Debut Era, 2008)

Demi Lovato (Debut Era, 2008) (rmvpe) (460 epochs)

(Reupload) With the restart of the AI Hub server, I thought it'd be a good chance to update one of my voice models. Please credit either my handle here or my Huggingface profile (magnus-son) if you use this model. Dataset 22 minutes of voice lines from the PC game "Nancy Drew: The Shattered Medallion" Extraction Method: rmvpe Sample Rate: 48kHz Epochs: 360 Huggingface: *Alt * NOTES: Recommended for speaking over singing. Transpose +10 for inferring male vocals. Search feature ratio of 1.0 works very well, although the model sometimes mistakes plosive "p" sounds for "k" sounds.

Nancy Drew (PC Games) (VA: Lani Minella)

Nancy Drew (PC Games) (VA: Lani Minella) (rmvpe) (360 epochs)

(Reupload) I LOVED these games as a kid and suddenly got the urge to make voice models for the core characters. *For the sake of brevityI'll post the individual samples in the replies.* Joni Savage (390 epochs): Santiago Rivera (400 epochs): Leslie Clark (340 epochs): Owen Lam (320 epochs): Laptrap (460 epochs): Notes: All models were trained by myself and use datasets from the PC games "6th Grade Adventures" and "Search & Solve Adventures". All speaking samples below use a search feature ratio of 1.0while singing samples use 0.5-0.6except for Joni's which I believe was still set to 1.0. Most also have the "protect voiceless consonants and breath sounds" set to 5.0 (disabled).

Jon Bellion (Mixtape Era)

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