WebKit

Introducing the LUMi (ExVOICE) (VOCALOID) RVCv2 RMVPE Model, powered by 300 epochs of training on 4m21s of exclusive ExVOICE samples. This advanced AI voice model excels at creating AI music and covers, offering a unique vocal range and style for musicians, producers, and hobbyists alike. Dive into the world of synthetic voices and explore endless possibilities for text-to-speech applications with this impressive tool. Start experimenting with our cutting-edge, free AI tools today and bring your creative visions to life.

DL: < Credit is optional, but appreciated This model was trained on 4m21s worth of audio from LUMi's exvoice samples.

LUMi (ExVOICE) (VOCALOID)

LUMi (ExVOICE) (VOCALOID) (RVCv2 RMVPE) 300 Epochs

mrjamiejames

datenshiluka

DL: < Credit is required if used. Please credit as mrjamiejames and link back to model. Well it's been a bit. Probably not going to be releasing any new models for the foreseeable future. Primarily due to a lack of interest. I had this sitting around for a while back when I was testing RMVPE. I'm no singer by any means so it's prolly new great sounding and it doesn't have a wide range, but ehh I just wanted to post one final thing before fading away back into the depths of IRL stuffs. The model does better in lower ranges and with softer songs. Enjoy I guess lol.

James Gantz (Myself)

James Gantz (Myself) (RVCv2 RMVPE) [400 Epochs]

DL: < Credit is optional, but appreciated. Oh boy this was a nightmare to train lmao. Since Colab is fucked I went through the effort to train this locally. My poor 1660ti Anyway, This here be Ruby. A Vocaloid that released in 2015, but actualy....It's Makune Hachi, an UTAUloid that predates Ruby. You see, Hachi and Ruby share the same voice provider and even better they also share the same tone so it makes them indistinguishable from each other. The dataset for this model consists of all of Hachi's UTAU VBs (Minus Fluffy) and it totals to be around almost 2 hours of audio. I could have 100% prolly trained longer, but again my 1660ti was having NONE of it lol. I think it sounds decent for the most part. Maybe someone will come in with a better one who knows and if that does become the same I'll gladly push people to use that one over this heh. Anyway hope yall enjoy! : )

RUBY/Makune Hachi (VOCALOID/Discontinued UTAU)

RUBY/Makune Hachi (VOCALOID/Discontinued UTAU) (RVCv2 RMVPE) 450 Epochs

DL: < Credit is appreciated, but optional This model was trained on 6m31s worth of audio from Nemu's exvoice samples.

Yumemi Nemu (ExVOICE) (VOCALOID)

Yumemi Nemu (ExVOICE) (VOCALOID) (RVCv2 RMVPE) 300 Epochs

DL: < Credit is optional, but appreciated Trained on around 4 minutes of audio from her V4 exvoice files Here's something to follow up my Unity-chan model from the other day while have the files to play around with heh. I'm aware there's already a Flower model available on this server, but that one was made using her raw dataset from Vocaloid. This one is made using her exvoice files. I wonder how this one compares to that one along with her commercial VBs. Anyway hope yall enjoy! : )

Flower(exVoice) (VOCALOID/CeVIO)

Flower(exVoice) (VOCALOID/CeVIO) (RVCv2 RMVPE) 300 Epochs

DL: < Credit is optional, but appreciated Trained on around 7 minutes of audio from her ExVoice files

LUMi (ExVOICE) (VOCALOID)

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