 Introducing the Emu Otori (Project Sekai) RVC v2 Model, now available after training for 300 epochs! This innovative AI voice model uses Retrieval-Based Voice Conversion (RVC) technology to bring your favorite virtual singer to life. With stunning accuracy, the Emu Otori model captures the unique characteristics of the original VA, Kino Hina. Use it to create AI covers or text-to-speech applications effortlessly.

As part of Project Sekai: Colorful Stage, this AI model embodies the spirit of collaboration between music and artificial intelligence. Ideally suited for producing AI covers or other creative projects, it boasts impressive versatility while preserving the distinctive qualities that make Emu Otori's voice stand out. Explore limitless possibilities using this sophisticated tool, backed by cutting-edge RVC technology.

Optimized through rigorous testing and fine-tuning, the Emu Otori model ensures seamless integration into your workflow. Whether you're crafting dynamic audio experiences or exploring new avenues for musical expression, our AI models deliver unmatched performance and dependability. To get started, simply sign up for access

this character is featured in project sekai: colorful stage. VA: Kino Hina this model is trained the same way as my nene model. trained with mangio-crepe and with 20 batch sizes drive

Emu Otori (Project Sekai)

Emu Otori (Project Sekai) (RVC v2) - 300 Epochs

sxndypz

this model now has a huggingface link, which is right below the drive link!

do u have the acapella, i wanna download it

buslert_

Для тех кто не понял, чтобы Эму звучала как Эму делайте тональность на +12 . Будет её голос

Синявка

jindocrackkkerrrrr

Ice Blue is a Cat

This is probably the most accurate model I’ve found!! It’s amazing

Kitty lover

AdelinalovesCATA

italian_espresso

♡harukiishibunn_

hello everyone im emu otori JAPPANESE LETER JAPANESE LETER

Victoria Agustina Barrera Lagos

Solar Productions

this character's featured in a show by rock!! mashumairesh!! anime, alongside with the stars!! anime and the fes a live game. trained on 7:27 minutes of her speaking from the first 8 episodes of the "show by rock!! mashumairesh!!" anime. (x-minus unwa mel-roformer v2 + mel karaoke (to remove some unneeded parts that the model picked up) + uvr de-reverb-de-echo + mel denoise + rx dialogue isolate + rx de-reverb (in some parts), resampled to 40k) cv. Misaki Watada pitch extraction: rmvpe steps: 17.2k batch size: 6 pretrain: klm 4.2 t4 / 40k huggingface:

Delmin (Show By Rock!! Mashumairesh!!)

Delmin (Show By Rock!! Mashumairesh!!) (RVC v2) - 540 Epochs

my first model trained in applio. this character's featured in a show by rock!! mashumairesh!! anime, alongside with the stars!! anime and the fes a live game. trained on 10 minutes of her speaking from the first 6 episodes of the "show by rock!! mashumairesh!!" anime. (x-minus unwa mel-roformer v1 + uvr de-reverb-de-echo + mel denoise + rx dialogue isolate + rx de-reverb (in some parts), resampled to 40k) cv. Hikaru Tohno pitch extraction: rmvpe steps: 15.4k batch size: 7 pretrain: klm 4.2 t4 / 40k huggingface:

Howan (Show By Rock!! Mashumairesh!!)

Howan (Show By Rock!! Mashumairesh!!) (RVC v2) - 420 Epochs

new month. new voice model. this voice model may sound a bit worse than the old dizzy model i made earlier trained on 16 minutes of audio data from her recent streams. (unwa's mel roformer + uvr de-reverb + rx de-click + de-plosive + eq) pitch extraction: rmvpe steps: 18.9k / 19k steps batch size: 7 pretrain: klm 4.1 / 32k huggingface:

Dizzy Dokuro (ACT2) (Phase Connect)

Dizzy Dokuro (ACT2) (Phase Connect) (RVC v2) - 380 Epochs

yay an another machikado mazoku voice model after a long time. trained on 7:30-ish minutes of audio data from the first 6 episodes of the 2nd season. (unwa's mel roformer + mel karaoke, resampled to 32k) cv. Tomoyo Takayanagi pitch extraction: rmvpe steps: 12.4k - 12.5k batch size: 6 pretrain: klm 4.1 / 32k huggingface:

Hinatsuki Mikan (Machikado Mazoku)

oh wow she has a wide range. i wonder if she could top my clara model trained on 16 minutes of audio data from her watchalong stream. (mel-roformer + uvr de-reverb + rx spectral de-noise + de-click + eq) pitch extraction: rmvpe steps: 16.2k - 16.3k batch size: 7 pretrain: klm 4.1 / 32k huggingface:

Enya Ignis (idol-EN)

Enya Ignis (idol-EN) (RVC v2) - 320 Epochs

HATSUNE MIKU V3 UPDATE You must credit if used. Change notes: Added index Trained on Harvest instead of PM Fixed model overall Remember, this post is always updating with updates for this model. See examples and showcases below You must credit if used.

FIXED HATSUNE MIKU V3 - VOCALOID [SCROLL DOWN FOR LINK]

FIXED HATSUNE MIKU V3 - VOCALOID (RVC2, 250 epochs) [SCROLL DOWN FOR LINK]

this character is featured in project sekai: colorful stage. this model is *now* trained with almost 20 minutes of data, with less of her screaming and crying and more of her talking from the map talk feature extraction: mangio-crepe hop length: 64 batch size: 15 VA: Sumi Tomomi Jiena drive

An Shiraishi (Project Sekai)

An Shiraishi (Project Sekai) (RVC v2) - 300 Epochs/7.8k Steps

Struggles a bit with songs with a higher range)

Mizuki Akiyama (Project Sekai Colorful Stage)

Mizuki Akiyama (Project Sekai Colorful Stage) (RVC V2) 350 Epochs

Toya Aoyagi (Project Sekai Colorful Stage)

Toya Aoyagi (Project Sekai Colorful Stage) (RVC V2) 400 Epochs

Made by me: (Please credit if use :3 @ yorukaiosei) Trained on Harvest with 37 minutes of dataset from Kanade Yoisaki's Alternate Vocals and Tomori Kusonoki's Album

Yoisaki Kanade [VA: Tomori Kusunoki] (Project Sekai)

Yoisaki Kanade [VA: Tomori Kusunoki] (Project Sekai) (RVC V2) 610 Epochs

yorukaiosei

- used approximately 10 minutes of singing - 1 minute of dialogue

Rui Kamishiro (Project Sekai)

Rui Kamishiro (Project Sekai) (RVC v2) 300 epochs

Tsukasa Tenma (Project Sekai Colorful Stage)

Tsukasa Tenma (Project Sekai Colorful Stage) (RVC V2) 400 Epochs

this model's trained on her talking and some of the samples from the dataset have their high range and low range so it should be good for her to sing and talk. VA: Machico trained with mangio-crepe and with 20 batch sizes. drive

Nene Kusanagi (Project Sekai)

Nene Kusanagi (Project Sekai) (RVC v2) - 300 Epochs

- Trained with mangio-crepe, 128 hop lenght - 10 minutes of dataset CREDIT AS humburgar IF YOU'RE GONNA USE THIS MODEL

Shinonome Ena (Project SEKAI)

Shinonome Ena (Project SEKAI) (RVC v2) 300 Epochs

humburgar

I hope you like her! She was trained with ~15 minutes of singing, so she's mainly suited for that. I want to make another model in the future with her talking voicelines. Also, if the Baka Mitai sample sounds different than usual, it's because I used Akiyama's cover, not Kiryu's :)

Emu Otori (Project Sekai)

Description

Add a comment

FIXED HATSUNE MIKU V3 - VOCALOID [SCROLL DOWN FOR LINK]

An Shiraishi (Project Sekai)

Mizuki Akiyama (Project Sekai Colorful Stage)

Toya Aoyagi (Project Sekai Colorful Stage)

Yoisaki Kanade [VA: Tomori Kusunoki] (Project Sekai)

Rui Kamishiro (Project Sekai)

Tsukasa Tenma (Project Sekai Colorful Stage)

Nene Kusanagi (Project Sekai)

Shinonome Ena (Project SEKAI)

Mafuyu Asahina (Project Sekai)

Saiba Momoi (Blue Archive)

Jungkook (BTS)

Sonic The Hedgehog [Roger Craig Smith]

ENHYPEN Heeseung

CaseOh

Villager (Minecraft)

FIXED HATSUNE MIKU V3 - VOCALOID [SCROLL DOWN FOR LINK]

Saiba Momoi (Blue Archive) (VA: Tokui Sora)

Satoru Gojo (JJK) [VA Yuichi Nakamura]

Fang [Brawl Stars]

Mortis [Brawl stars]

Melodie - (Brawl Stars) - (English & Korean Voices)

Takanashi Hoshino (from Blue Archive)

Sas (Counter-strike global offensive)

Edgar [Brawl Stars]

SpongeBob SquarePants (Talking And Singing)

Ariana Grande AI

JP Furina / Focalors CV: Inori Minase (Genshin Impact)

Surge (brawl Stars)

JENNIE of BLACKPINK [Strong Ver.]