Name: Mem Cho(Oshi No Ko)| Jp, En, Ru | AI RVC Model
Brand: Weights
Rating: 4.2 (19 reviews)

WebKit

 Introducing our latest AI Voice Model, Mem Cho(Oshi No Ko) RVC v2, powered by Retrieval-Based Voice Conversion (RVC) technology. This advanced model has been trained on a diverse 1.5-hour dataset consisting of one season of a popular anime series and the Mem YouTube show, making it well-equipped to handle both Japanese and English accents as well as Russian language commands. With over 3000 epochs of training using Sl17, this model boasts exceptional precision and versatility.

The Mem Cho(Oshi No Ko) RVC v2 model stands out due to its ability to generate high-quality AI music and create impressive AI covers of your favorite tunes. Its sophisticated programming allows for unparalleled accuracy and seamless integration into various applications. Whether you're interested in text-to-speech conversion or generating custom audio files, our cutting-edge AI technology is here to help.

At Weights, we offer a range of powerful yet easy-to-use AI tools designed to simplify your workflow and enhance creativity. Don't miss out on the opportunity to explore all that the Mem Cho(Osh

- Trained on rmvpe about 1.5 hour dataset from 1 season of anime and Mem YouTube show - 2960 epochs, Sl177600 - Speech to speech, txt to speech, can sing - Works best on rmvpe - Also I'm working on mangio crepe version - Credits are much appreciated

Mem Cho(Oshi No Ko)| Jp, En, Ru

Mem Cho(Oshi No Ko)|RVC v2| 3000 Epochs| rmvpe| Jp, En, Ru

mustar22

https://huggingface.co/MUSTAR/MemChoRVC/resolve/main/MEM_V1.zip

omochabox

<a:MemchoPeaceOut:1129579734959923250> <:MemchoPleased:866863178380869662>

thanks for explaining. Tomorrow I'll send here another 3 model to merge. I figured it out that if you merge models in different languages, the voice begins recognize speech better. And the final result doing better job rather than just one.

Mem Cho(Oshi No Ko)|RVC v2| 3600 Epochs| rmvpe| Jp, En, Ru

Updated version of Mem is ready to rock. Now it will do the job better

Mem Cho(Oshi No Ko)|RVC v2| 3600 Epochs| rmvpe/mangio crepe| Jp, En, Ru

https://discord.com/channels/1089076875999072296/1153425789530939512

⭐ 𝓚𝓾𝓶𝓪 ⭐ ʕっ•ᴥ•ʔっ

42 min dataset 10 eps 2550 steps Pretrain: RIN_E3 download :

Gawr Gura U (preview) / 10 eps / RIN_E3

10 min dataset of bad quality speech of Zuck 10 eps 560 steps Pretrain: RIN_E3 download :

Mark Zuckerberg U (preview) \ 10 eps \ RIN_E3 pretrain

1- PunkaLopi: 1 hr 1 min 2- Cy Yu: dataset 1 hr 3 min 3- Mem Cho: dataset 1 hr 5 min 4- Kobo: dataset 57 min 5- Hoshi: dataset 5 min 6- Suisei: dataset 14 hrs team: @aleks_29 @simplcup @Blaise @.pussypunisher pretrain ### hrs epochs: 25 Steps: 520464 (still in process of training)

Demo preview of models made on ##### dataset

Ladies and gentlemen we got em!! Introducing the Snowie the Russian Pretrain!! !! 1) Snowie V1.2 48k (based on EngPretrain) 2) Snowie V2.2 40k (based on EngPretrain) 3) Snowie V2.2 40k (no pretrain) (in addition: English based Snowie kinda does well in English) Extraction algorithm (RMVPE) A mini pretrain based on Russian language. *Improves Russian. "" . Target sample rate: 40k and 48k Dataset duration \ : 21 hrs (cristal clean) Total epoch count \ : 2) 450 ep, 3) 400 ep, 1) 350 ep How to use: 1 Download the G and D pth files 2 Inside the Fork find folder named: "pretrained_v2" 3 Next paste g and d files direction into "Load pre-trained base model G/D path:" 4 Ur ready to Rock! : 1 G D pth 2 pretrained_v2 3 4 ! Feel free to experiment and compare pretrains I appreciate credits, so kindly acknowledge the creator when sharing results.

"Snowie \ Снежок" Russian Pretrain. (3 versions)

(3 versions) VJp: VRu: ULTIMATE : First of all..... pain Aqua: I don't care. My favorite always was and always will be Ai . AI_Hoshino from Oshi no ko Dataset_duration/Epochs: V_Japanese-3hrs/5000 ep V_Russian-2hrs/500 ep V_Ultimate-4hrs/1000 ep Extractor: RMVPE Pretrain: custom made mini RuPretrain "Snowie" and classic English pretrain *Improoves Russian, without affecting English. Languages: English, Russian, Japanese Voice to voice, TTS, Scr of Tensorboard$ below. Few versions included. Feel free to compare. We appreciate credits, so kindly acknowledge the creator when sharing results.

Mem Cho(Oshi No Ko)| Jp, En, Ru

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