WebKit

Introducing the Lightning McQueen RVC v2 500 Epochs model, a unique Tacotron2 text-to-speech creation based on the beloved character from the 'Cars' series on The Cool Videos YouTube channel. Originally created during the Uberduck era, this voice gained popularity through wayneradiotv's use. With improved quality, generate AI covers and text-to-speech with this whimsical, one-of-a-kind voice, absolutely free. Perfect for adding fun audio twists to your projects.

He's finally here, it is him, Lightning McQueen, from the new movie, Cars! ...or perhaps you might know this voice better as a certain gnome from a certain streamer, which in that case... *you have failed the challenge.* This is the bizarre McQueen as portrayed in the "Cars" video series from "The Cool Videos" friend-group-shitpost-type channel on YouTube. I turned it into a Tacotron2 text-to-speech model back in the Uberduck days, and *wayneradiotv* used it, making it famous for different reasons. (Feel free to let him know about this model by the way lol) *Now, you too can say the darndest things with this silly voice in better quality at no additional cost :)*

Lightning McQueen (The Cool Videos) / Gnome Chompski (wayneradiotv)

Lightning McQueen (The Cool Videos) / Gnome Chompski (wayneradiotv) RVC v2 500 Epochs

All "Cars" videos + similar stuff from the same people for anyone curious/intrigued/Disgusted: <https://www.youtube.com/playlist?list=PLyhIO-4tEMIB_Hoc1zC0jWsAaStDL9WJd>

wayneradiotv talking about finding the voice on Uberduck → <https://youtu.be/GRj9iQjyKoQ?t=1265>

.supers.

I wonder if you were to put the gnome clips through this model they'd sound "enhanced"... lazy to test that right now tho

What pitch should I put the voice at to make it sound similar to Gnome Chompski?

Logan Meek

@Logan Meek just raise an octave (+12) if you're using lower pitch samples

Supposed to be my "ultimate model" with a lot more epochs than this (until overtraining happens) given the extensive dataset of nearly 1 hour of clean audio from different Winnie the Pooh games, but because of... problems with the colab I don't think I can continue training this one further, I may have to retrain it from the beginning. THAT SAID, the model is already pretty good for speaking, as you will notice with the first three samples, which originate from a Pooh game where, I guess they did *not* have the budget to get the real Laurie Main to do narration, and got a "soundalike" instead. It's good enough though that, through this model, it sounds indistinguishable from the actual voice now. < OLD version (150 Epochs): <

Winnie the Pooh Narrator (Laurie Main)

Winnie the Pooh Narrator (Laurie Main) RVC v2 300 Epochs

tee hee Watch out, Neco Arc, *you got some competition* VA: Satomi Krogi Trained on available clean audio from Klonoa 2 (22050Hz)

Tat (Klonoa 2: Lunatea's Veil)

Tat (Klonoa 2: Lunatea's Veil) RVC v2 300 Epochs

A snowball in July...? trained on 10 minutes of data, the YTPMV legend voiced by Wally Wingert who also voiced many other characters in television such as Jon Arbuckle and the red Tallest from Invader Zim.

Lightning McQueen (The Cool Videos) / Gnome Chompski (wayneradiotv)

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