WebKit

 Introducing the advanced Usher [So-Vits 33.6K] [RVC 1000 UPDATE], an innovative AI Voice Model powered by Retrieval-Based Voice Conversion (RVC) technology. This state-of-the-art model captures the unique vocal characteristics of renowned singer Usher, enabling you to generate remarkably authentic-sounding AI covers and text-to-speech outputs. With its exceptional 33.6K quality and recent RVC 1000 update, the UsherCOMEHU model delivers unparalleled performance in creating AI music and speech.

Developed using a diverse rap dataset, this sophisticated model brings Usher's distinctive style and rhythm into your projects. Ideal for musicians, producers, and developers seeking high-quality AI voices, the Usher [So-Vits 33.6K] [RVC 100

Link [So-Vits-SVC]: < Link [RVC]: < Modelname: UsherCOMEHU (index file added) The dataset included rap acapellas too but i havent test it at all Credits to me

Usher [So-Vits 33.6K]

Usher [So-Vits 33.6K] [RVC 1000 UPDATE]

comehu

Example RVC 600 epoch (mixed version on youtube): https://www.youtube.com/watch?v=T_dYexq6mTg

Usher [So-Vits 33.6K] [RVC 1000 Epoch UPDATE]

Titan model trained on 17 minutes of clean acapellas, manually sorted by me COMEHU. Extra keywords: Jadiel El Tsunami / Jadiel El Leon Del Area Sur

Jadiel El Incomparable (TITAN PRETRAIN)

Jadiel El Incomparable v2 (TITAN PRETRAIN)

New Tinashe model based on BB/Ang3l and Joyride albums. Used more acapellas from Joyride. Used TITAN pretrained model to make this Tinashe model

Tinashe (mainly Joyride era) TITAN MODEL

Retrained Aaliyah model on TITAN pretrain. Includes acapellas from Aaliyah's red album and I Care 4 U this TITAN retrain includes a new acapella "Enough Said" for more accurate low notes interpretation this change also makes the high ends more sharper. Both models rely on isolated acapellas and some official studio acapellas i found and cleaned

Aaliyah (Aaliyah's era) TITAN HQ MODEL

Unserious model i used TITAN pretrained model to make this. I used audio samples from the recreation made by scientists by producing a 3D-printed voice box based on Nesyamun's vocal tract which were really bad and were turned into a meme. Lowering the "Search Feature Ratio" to zero makes the model pronunciate the words a lil more defined by default this model can only produce just one vowel.

Priest Nesyamun (3,000-year-old mummy voice recreation meme)

trained on dry acapellas and manually cleaned by me. Dataset is from 2016 to 2022 including "Lift Me Up" Credits to me

Usher [So-Vits 33.6K]

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