Name: Mai Sakurajima (Rascal Does Not Dream) | AI RVC Model
Brand: Weights
Rating: 4.0 (43 reviews)

WebKit

 Introducing the Mai Sakurajima (Rascal Does Not Dream) RVC v2 - an advanced AI Voice Model powered by Retrieval-Based Voice Conversion (RVC) technology. This innovative model captures the distinct voice characteristics of Mai Sakurajima, as played by the talented Asami Seto in the popular anime series Rascal Does Not Dream of Bunny Girl Senpai. With the ability to create stunningly realistic AI covers and text-to-speech applications, the Mai Sakurajima (Rascal Does Not Dream) RVC v2 sets itself apart as a premier choice for developers and creators alike. Experience life-like performances using our state-of-the-art AI technology, designed specifically for those seeking high-quality audio reproductions. Try out our free AI tools today at Weights and revolutionize your creative process! Don't miss out on crafting remarkable AI covers and speeches featuring the iconic tones of Mai Sakurajima. Begin exploring now – take advantage of cutting-edge RVC Model capabilities within the realm of AI music and speech generation.

Mai Sakurajima from the Rascal Does Not Dream of Bunny Girl Senpai anime, voice acted by Asami Seto. Trained by me on 5 minutes of audio from season 1. Credit if used please. RMVPE at 200 epochs < RMVPE at 325 epochs < Mangio-Crepe at 400 epochs < More BGS characters coming soon

Mai Sakurajima (Rascal Does Not Dream)

Mai Sakurajima (Rascal Does Not Dream) (RVC v2)

july.lion

# <:MaiSmug:500773520339632149>
Mai Sakurajima from the Rascal Does Not Dream of Bunny Girl Senpai anime, voice acted by Asami Seto.
Trained by me on 5 minutes of audio from season 1. Credit if used please.

**RMVPE** at **200 epochs**
<https://huggingface.co/JuLY-LION/Mai-San/resolve/main/Mai-San-v5_e200_s3400.zip>
**RMVPE** at **325 epochs**
<https://huggingface.co/JuLY-LION/Mai-San/resolve/main/Mai-San-v5_e325_s5525.zip>
**Mangio-Crepe** at **400 epochs**
<https://huggingface.co/JuLY-LION/Mai-San/resolve/main/Mai-San-v4-e400_s0.zip>

More BGS characters coming soon™️

Why did I include 3 models? Mangio-Crepe was created first because I wasn't aware there was anything better yet, and I only had 3½-4 minutes of sample audio. As far as the training graph during RMVPE was concerned, it didn't start overtraining until after 325 epochs, but feel free to try 200 epochs if it sounds overtrained anyways.

unknownplayrs

ah alright, if u don't mind me asking, what batch size did u use?

but it sounds good, glad someone made an updated version of this

The game announcer from Cobalt (+Cobalt WASD) by Mojang and Oxeye Game Studios. Dataset extracted right from the game, roughly 90 seconds of audio. Training by me. Mangio-Crepe 150 epochs 1,050 steps hop length of 96 batch size of 10

Cobalt (WASD) Game Announcer

Cobalt (WASD) Game Announcer (RVC v2) (150 epochs)

This is Markiplier during his playthrough of the game Inscryption, and his voice acting of Leshy. Gonna be honest, this one didn't turn out as well as I hoped. Markiplier created a certain style that RVC couldn't fully capture. Dataset and training by me. Dataset length is 3:20. Mangio-Crepe. Hugging Face:

Leshiplier

Leshiplier (RVC v2)

Combat Arms is an FPS game from the 2010's. This is the announcer for the gamemode Quarantine Regen. Trained by me. Made for the Halloween contest on this server but I didn't place  (probably because I only had 16 seconds of sample data, but I still think it came out well) Huggingface: < I don't remember the pitch extraction method but it was probably Mangio-Crepe.

Combat Arms Quarantine Regen Announcer

Combat Arms Quarantine Regen Announcer (RVC v2) (575 epochs)

Ernst Zimmerman, voiced by Keith Silverstein, in the English dub of the EIGHTY-SIX anime. Dataset and training done by me. Dataset length is about 2 minutes 50 seconds. Huggingface: Mangio-Crepe at 250 epochs: Oh by the way, 86 is a phenomenal show. S1-P2 ranked in the top 50 anime of all time.

Ernst Zimmerman: EIGHTY-SIX

Ernst Zimmerman: EIGHTY-SIX (RVC v2) (250 epochs)

The narrator behind MythBusters, voiced by Robert Lee. Trained by azumangadaio_h. Commissioned by me. I helped with the dataset a little. Yes, Pepsi did this model. I commissioned this from her well before the server nuking. *Don't shoot the messenger.* Huggingface: < Mangio-Crepe, but Pepsi never mentioned the epoch count. Dataset is at least 3 minutes.

Mai Sakurajima (Rascal Does Not Dream)

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