WebKit

Introducing the Oki (Oskar Kamiński) Polish Rapper RVC Model, retrained from the RVC V1 with an improved dataset including songs from 2022-2023. This model uses Crepe V2 and features a 1.0 feature ratio. Get ready to create amazing AI covers and text-to-speech using this powerful tool, all thanks to its creator who provides updates and improvements along the way. Don't forget to give them credit on YouTube and TikTok.

Retrained from the RVC V1 one! Made by me Same Dataset Trained with crepe V2 16th RVC V2 Model. USE FEATURE RATIO TO 1.0 ALWAYS. Credit to me on Youtube & TikTok. UPDATED: Retrained with Mangio-Crepe Extraction & Improved Dataset (+ Songs that i Forgot) (2022-2023 Songs Only)

Oki (Oskar Kamiński) (Polish Rapper)

CrimsonZockt

Oki (Oskar Kamiński) (RVC V2) 300 Epochs

I think i overtrain Doda, 250 Epochs for her.

All of my RVC V1 models have been retrain to RVC V2 models! You can use covers for a good quality!
Huggingface link Up!

https://huggingface.co/CrimsonZockt/RVC-Model/resolve/main/OkiV2.zip

Join the server if you are interested in Polish AI Models.
https://discord.gg/9vkWFbpeT2

Retrained with Mangio Crepe Extraction & Improved Dataset.

Oki (Oskar Kamiński) (Polish) (RVC V2) 300 Epochs

Oki (Oskar Kamiński) (Polish Rapper) (RVC V2) 300 Epochs

Trained on Applio Locally OG Pretrain, Sample Rate 48k, contentvec Embedder, HifiGAN Vocoder, Batch Size 4, FP16 Precision, And 18 Minutes Dataset (Album) Index Ratio 1.0 Recommended Credit to me On Youtube (@FuturisticLabelAI) & Tiktok (@futuristiclabelai)

Julia Posnik (Egoistka Era)

Julia Posnik (Egoistka Era) (RVC V2) (RMVPE) (350 Epochs)

Trained on Applio Locally I decided to on the version v3! This old one must be promise OG Pretrain, Sample Rate 48k, contentvec Embedder, HifiGAN Vocoder, Batch Size 4, FP16 Precision, And 9 Minutes Dataset (All Singles and so on) Index Ratio 1.0 Recommended Credit to me On Youtube (@FuturisticLabelAI) & Tiktok (@futuristiclabelai)

Blanka Stajkow (All-Round)

Blanka Stajkow (All-Round) (RVC V2) (RMVPE) (300 Epochs)

Trained on Applio Locally One Song "SMUTNO MI" had cutoff 16hz so, I used AudioSR. I don't use AudioSR because sucks! OG Pretrain, Sample Rate 48k, contentvec Embedder, HifiGAN Vocoder, Batch Size 4, FP16 Precision, And 20 Minutes Dataset (Only Songs) Index Ratio 1.0 Recommended Credit to me On Youtube (@FuturisticLabelAI) & Tiktok (@futuristiclabelai)

Julita Maria Kosy (Marie) (Babygirl Era)

Julita Maria Kosy (Marie) (Babygirl Era) (RVC V2) (RMVPE) (350 Epochs)

Trained on Applio Locally TITAN Pretrain, Sample Rate 40k, contentvec Embedder, Batch Size 4, FP16 Precision, And 24 Minutes Dataset (Mniejszeodtrzech Album) Index Ratio 1.0 Recommended Credit to me On Youtube (@FuturisticLabelAI) & Tiktok (@futuristiclabelai)

Magda Bereda (Mniejszeodtrzech Era)

Magda Bereda (Mniejszeodtrzech Era) (RVC V2) (RMVPE) (550 Epochs)

REPOSTED Trained on Applio Locally TITAN Pretrain, Sample Rate 40k, contentvec Embedder, Batch Size 4, FP16 Precision, And 11 Minutes Dataset (nitki Album) Index Ratio 1.0 Recommended Credit to me On Youtube (@FuturisticLabelAI) & Tiktok (@futuristiclabelai)

Oki (Oskar Kamiński) (Polish Rapper)

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