Name: Announcer - PORTAL: Reloaded | AI RVC Model
Brand: Weights
Rating: 4.2 (1 reviews)

WebKit

Introducing the "Announcer - PORTAL: Reloaded" RVC Model, crafted by Weights, an advanced AI Voice Model perfected over 630 epochs to deliver premium-quality voice conversion. Developed using 4 minutes and 42 seconds of intricately curated data from the Portal Reloaded mod, this model utilizes sophisticated preprocessing like RX de-ess, mouth de-click, resampling, and EQ noise gate for incredibly crisp and clear outputs. With pitch extraction powered by Mangio-Crepe, this RVC Model excels in generating lifelike speech, making it ideal for text-to-speech applications. While it may have an unconventional edge for singing, its true strength lies in delivering articulate and engaging AI music covers and immersive voiceovers. Experience the future of AI music and voice conversion with Weights. Create AI Covers or enhance your projects with our free AI tools now!

trained on 4 minutes and 42 seconds of data from the portal reloaded mod. (rx de-ess, mouth de-click, resample + eq, noise gate) - dataset provided by: @Eko(Childish AI Hubber) cv. Harry101UK pitch extraction: mangio-crepe (applio crepe) steps: 5.4k batch size: 6 pretrain: original v2 / 32k huggingface: model is kinda strange at singing, recommended use for speech. a index ratio of 0.5 - 0.75 is recommended. please don't forget to credit me when you use this model. please don't reupload any of my models without my permission, otherwise i'll report them to be taken down.

Announcer - PORTAL: Reloaded

Announcer - PORTAL: Reloaded (RVC V2, mangio-crepe, 630 epochs)

Y O G U R T

: 9:37 minutes, ripped from youtube (opus), to be specific, his oldest videos. gpt epochs: 22 sovits epochs: 20 sample rate: 32k recommended to use slice by every english punct. don't forget to credit me when using this model or any of my models - read the license file of the models' repo for more information. audio and transcripts inside zip!

ThatGuyWithAVoice

ThatGuyWithAVoice (GPT-SoVITS V2) (20-22 epochs)

model trained on 24 minutes of dataset ripped from youtube. please don't forget to credit me when you use my models, and don't reupload any of my models without my permission. read the license file of my models repository for more info.

Sonic the Hedgehog (Sonic & The Black Knight)

trained on a 1-hour and 2-minute dataset. please don't forget to credit me when you use this model, and don't reupload any of my models without my permission. read the license file of my models repository for more info.

Wheatley (PORTAL 2)

model trained on 3-minutes and 20-seconds of dataset. please don't forget to credit me when you use this model, and don't reupload any of my models without my permission. read the license file of my models repository for more info.

Meeple (Brawl Stars)

link - a 19-minute and 27-second dataset ripped from almost all songs of the "A Very 1999 Christmas" album. it could have been better, but i didn't want to waste my work cleaning the dataset. *this is a retrain of balbal's 1234 epoch jschlatt model btw* batch size: 8 klm 4.2 pretrain sampled at 32k please don't forget to credit me when you use this model, and don't reupload any of my models without my permission. read the license file of my models repository for more info.

Announcer - PORTAL: Reloaded

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