Name: Nyx (Hades I) | AI RVC Model
Brand: Weights
Rating: 4.2 (1 reviews)

WebKit

Experience the enchanting depths of "Nyx (Hades I) [Mommy voice]," an advanced RVC Model perfect for AI Music enthusiasts and developers. Born from the mystical essence of Mother Night herself, this AI Voice Model beautifully captures the nurturing yet powerful voice of Nyx from the acclaimed game Hades. Developed over 347 epochs with the RMVPE 32K framework, Nyx's voice model delivers an enthralling audio experience despite minor harmonic compromises due to reverse reverb removal. Recommended pitch adjustments range from -1 to -2 for optimal inferencing. Unlock the potential to create AI Covers and enhance your Text-to-Speech projects with Nyx's unique vocal characteristics. Try Weights' free AI tools today and revolutionize your audio creations!

**"Mother Night we like to call her though rare is the occasion when she dotes on us."* It was difficult to remove the reverse reverb from her voice lines so the harmonics is slightly damaged. I suggest a pitch of -1 to -2 for inferencing if the default isn't enough

Nyx (Hades I)

Cauthess

dataset length was 2 minutes so the accent is close enough

For inferencing, you can reverse the audio in audacity, add in 50% of Maes Howe reverb to the audio (from Songpeel Reverb adder), unreverse the audio and it'll give the ethereal effect. This is a more of a f2p version of how the devs apply their processing techniques

Illyana Rasputina, mutant Sorcerer, brother of Colossus and ruler of Limbo Try to play around with the pitch depending on the female voice for inferencing! Pitch 12 isn't guaranteed to sound good so lower it to 3 for high pitch voices

Magik (Marvel Rivals)

Magik (Marvel Rivals) (RVC V2) (RMVPE) (288 Epochs)

It's the girl from Fortnite! Gwen only had a single animation featuring her voice "the Gwenpool Holiday Special". I used elevenlabs to create more lines so it's more usable. VA: Brenna Hines

Gwenpool (Marvel TLDR / Fortnite)

She's the distant cousin of Zagreus! The model works better with a british accent. 2 minute dataset made using only RX11 Dialogue Isolate

Artemis (Hades I)

Artemis (Hades I) (RVC V2) (438 Epochs) (RMVPE 32k)

This is every roleplayer's favorite lil' guy, James Randal! This may be the longest model I've done yet since I decided to challenge myself to create a high quality voice model meant for realtime. I have went through hours of Spaceboy clips until his mic quality started deteriorating. I only ask that you test this thoroughly for yourself with the realtime voice changer before you take this to RP since I can't control what you do For inferencing, you can use a 0 pitch for both male and female - minutes of clean audio - Pretain: Titan (I meant to do KLM 4.1, but it slipped my mind) - DAW used: Izotope RX11 Might include fp32 or full precision training - Model fusion used to merge the best G & mel cpkt - Synced the graphs on the tensorboard Model

James Randal - NoPixel RP (Spaceboy)

"HELLO". Trained on 7 minute and 30 seconds of audio voicelines from Multiversus. I included his screams/yelling to help him handle high pitches. For inferencing you can use a pitch of 0 for male and female. VA: Austen Moret

Nyx (Hades I)

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