WebKit

V4. Model trained using isolated tracks from across Dave’s career. Vocals taken from select songs from the albums The Colour And The Shape, There Is Nothing Left To Lose, One By One, In Your Honor, Echoes Silence Patience & Grace, 01050525, and Medicine At Midnight by Foo Fighters as well as their 2022 metal album released as Dream Widow. Vocals also taken from the songs “In Bloom” by Nirvana, “Goodbye Lament” by Tony Iommi, and “Beelzeboss (The Final Showdown)” by Tenacious D. All stems have had noise, echo, and reverb removed and were compressed + normalized to -1 dB for the sake of achieving an even & consistent sound. V4 was trained on a longer dataset than previous versions.

Dave Grohl

Rampaging Moose

V1. Model trained using Hetfield’s isolated vocal tracks from the entirety of Metallica’s 2008 album Death Magnetic as well as the 2011 follow-up EP Beyond Magnetic. I tried removing as much noise as possible from the stems but due to Death Magnetic’s notoriously brickwalled production there was only so much I could do. All stems have had echo and reverb removed and were compressed + normalized to -1 dB for the sake of achieving an even & consistent sound.

James Hetfield (Death Magnetic + Beyond Magnetic)

V1. Model trained using Corey’s isolated vocals from the entirety of Slipknot’s performance at Download Festival 2009 which can be found as a bonus disc on the band’s 2012 greatest hits compilation Antennas To Hell. All stems have had noise, echo, and reverb removed and were compressed + normalized to -1 dB for the sake of achieving an even & consistent sound.

Corey Taylor (Live at Download Festival 2009)

V2. Model trained using isolated vocal tracks from across Cantrell’s career. Vocals taken from the albums Alice In Chains, MTV Unplugged, Black Gives Way To Blue, and Rainier Fog by Alice In Chains. Vocals also taken from the albums Boggy Depot, Degradation Trip, and I Want Blood by Jerry Cantrell. All stems have had noise, echo, and reverb removed and were compressed + normalized to -1 dB for the sake of achieving an even & consistent sound. Nothing but love and respect for Jerry Cantrell!

Jerry Cantrell

V6. With previous attempts at making a Dax model I primarily focused on using his vocals from Acid Bath. Unfortunately I was unable to get great results due to the fact that his vocals on the Acid Bath albums tended to be heavily layered and effects heavy. So I decided to try a different approach. V6 was trained on Dax’s vocals from the entirety of his 2007 solo album We Sing Of Only Blood Or Love. Vocals also taken from the albums When The Kite String Pops, Paegan Terrorism Tactics, and Demos 1993-1996 by Acid Bath. Vocals also taken from the song “Stop, I’m Already Dead” by Deadboy & The Bunnymen. All stems have had noise, echo, and reverb remember and were compressed + normalized to -1 dB for the sake of achieving an even & consistent sound.

Dax Riggs

V1. Experimental model trained on Dave Grohl’s vocals from the song “Beelzeboss (The Final Showdown)” by Tenacious D as well as a small amount of his more aggressive vocals from a few songs by Foo Fighters/Dream Widow.

Dave Grohl

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