WebKit

 Introducing the Madobe Ai v2, a state-of-the-art Retrieval-Based Voice Conversion (RVC) Model inspired by Windows 8's operating system character, Madobe Ai. This advanced AI voice model has been trained using a vast dataset of various Madobe Ai themes, including a special edition, totaling 3 minutes and 52 seconds of high-quality audio. With over 500 epochs of refinement, the Madobe Ai v2 delivers exceptional naturalness and clarity, making it perfect for use in text-to-speech applications and creating AI covers.

As an upgraded version of its predecessor, the Madobe Ai v2 boasts improved performance, offering more accurate and expressive vocal capabilities. Developed utilizing the latest advances in artificial intelligence music technology, this innovative RVC Model ensures seamless integration into your projects while delivering unparalleled results. Whether you need lifelike narration or captivating musical performances powered by AI, the Madobe Ai v2 promises to deliver.

To start exploring the endless possibilities offered by the Madobe Ai v2 and other cutting-edge AI models, check out our

with 3 minutes and 52 seconds of dataset from various Madobe Ai Themes (including the special editions) + Windows 10 Pro DSP Limited Pack Voice Drama [WAY BETTER than the previous one I've made] Normalized Voiced by Ai Nonaka 8k Steps RVMPE/64 Hop-Length 40k Target Sample Batch Size 7 NOTE: Didn't include the 2 songs singed by Madobe Ai and Yuu because the voices sound a bit muffled compared to their talking voice

Madobe Ai (Windows 8's OS-tan)

Madobe Ai v2 (Windows 8's OS-tan)

SyberG

🔗Link: https://huggingface.co/SyberGen/OStans-VModels/resolve/main/MadobeAi-v2.zip?download=true
> 🚆Trained with 3 minutes and 52 seconds of dataset from various Madobe Ai Themes (including the special editions) + Windows 10 Pro DSP Limited Pack Voice Drama [WAY BETTER than the previous one I've made]
> ✨Normalized
> 🎙️Voiced by Ai Nonaka
> 
> 👟 8k Steps
> 🔊 RVMPE/64 Hop-Length
> 🎯 40k Target Sample
> 💼 Batch Size 7
> 
> NOTE: Didn't include the 2 songs singed by Madobe Ai and Yuu because the voices sound a bit muffled compared to their talking voice

200 Epochs + RVC > Trained with 2 minutes and 27 seconds of dataset > Original Pretrain from Applio > Noise Reduced + Reverb Reduction doesn't work > 1.8k Steps > RVMPE/64 Hop-Length > 40k Target Sample > Batch Size 8

President of Galaxy [Kukoo Kitchen]

430 Epochs + RVC > Trained with 2 minutes and 13 seconds of dataset > Original Pretrain from Applio > Noise Reduced + Reverb Reduction doesn't work > 4.3k Steps > RVMPE/64 Hop-Length > 40k Target Sample > Batch Size 8

Jessica [Kukoo Kitchen]

250 Epochs + RVC > Trained with 2 minutes and 51 seconds of dataset > Original Pretrain from Applio > Noise Reduced + Reverb Reduction doesn't work > 2,750 Steps > RVMPE/64 Hop-Length > 40k Target Sample > Batch Size 8

Matt [Kukoo Kitchen]

370 Epochs + RVC > Trained with 2 minutes and 46 seconds of dataset > Original Pretrain from Applio > Noise Reduced + Reverb Reduction doesn't work > 4,440 Steps > RVMPE/64 Hop-Length > 40k Target Sample > Batch Size 8

Chill-Li [Kukoo Kitchen]

200 Epochs > Trained with 9 minutes and 23 seconds of dataset > Original Pretrain from Applio > Noise Reduced > Voiced by Chuck Huber > 6k Steps > RVMPE/64 Hop-Length > 40k Target Sample > Batch Size 8

Madobe Ai (Windows 8's OS-tan)

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