Super Smash Bros. Announcer - Master Hand (from SSBU)
Non-Voice / OtherFictionalRVC v2English

📣

From the creator

"I have a patreon where you can support my RVC stuff (colabs guides models etc) and also get access to direct tech support on how to use the models: Trained on Harvest due to the quality of source clips not being so good. Besides that it came out fantastic after you apply a special delay preset on the vocals in an audio editor (I removed Reverb + Echo when training for quality reasons)"

Description

Introducing the Super Smash Bros. Announcer – Master Hand (from SSBU) RVC v2 Model, a cutting-edge artificial intelligence voice model developed by Weights. With 400 epochs of advanced training, this RVC Model delivers highly realistic and accurate vocal simulations of the iconic Master Hand announcer from Super Smash Bros. Ultimate. Utilizing Retrieval-Based Voice Conversion (RVC) technology, our AI music tool creates seamless and authentic-sounding covers, making it perfect for musicians, producers, and enthusiasts looking to explore new creative frontiers. In addition to its impressive capabilities as an AI Cover Maker, this versatile model can also generate high-quality text-to-speech outputs. Join the ranks of satisfied supporters on Patreon who already enjoy exclusive access to collaborations, guides, and additional models. Unleash your creativity today—try out our free AI Tools and start creating AI Covers and Text-to-Speech like never before!

See more

Weekly Metrics

Samples

New
Classic

Pitch

0

1

Male

Singing

🇺🇸 English

2

Female

Singing

🇺🇸 English

3

Female

Singing (Dry)

🇺🇸 English

4

Female

Singing (High)

🇺🇸 English

5

Male

Singing 2

🇺🇸 English

6

Male

Singing (Dry)

🇺🇸 English

7

Male

Singing (Dry, High)

🇺🇸 English

Comments

Apply this delay preset twice (at first it was -16db once? but I think this is better) to get the same echo effect. You may wanna play with these settings manually I couldnt get it 100% but I got it super close. It's worth experimenting with on your own if you're not satisfied with this.

image.png

(Standard reverb settings)

Super Smash Bros. Announcer / Master Hand (from SSBU) (RVC v2) 400 Epochs

Alternatively you can try doing this then this for a less aggressive but maybe 'better sounding' reverb?

image.png
image.png

(Alt reverb settings)

Wtf this is the 2 and a half minute dataset model. I thought I was training on 6 minutes but this whole time it was the 'trimmed to best quality' dataset the whole time and it still sounds great

this is really good. better than my announcer

If you want you can try to do a retrain of that one using De-Echo De-Reverb in UVR (what I used for this)

the output for my dataset looked like this after that so I trained with harvest and I got phenomenal results

i intentionally left the echo but i might try it

oh yea I also had a version of the announcer before this w/ echo as a test and it sounded pretty awful lol

This is how the model would've sounded without echo removal:

i havent used that model before but when i do i get float division by zero error.

you mean this one?

oh wait you mean in UVR

if you're running locally are you running on a PC w/ a GPU?

yeah i have an rtx 2060 super

you downloaded it from the download center?

yeah

under that glove bro got the hands of roddy rich

Master Hand - Don't Smash Me Now

sounds better because thats how he talks

you just need to do it in post instead

Super Smash Bros. Announcer - Master Hand (from SSBU) (RVC v2) 400 Epochs

Thank you Kalo 🙏

I CANT SEE ME LOVING NOBODY BUT YOU

THIS GOES HARD

SEND ME THE VOCALS NOW

finally did it

beautiful

Add a comment

Post