Amo Awesome Art
Song with my very own original lyrics, used to test out the suno ai service, I liked the final result and wished to see how it would sound as Pinkie, sadly just plugin in voice convertors was not possible as suno is adding some kind of auto tune on top of random duvet and reverb voice effect, so I had to first convert it to "proper Pinkie Pie" with the talknet, than run that through RVC, replace about 1/3 of the words as the pronunciation and pitch keep coming out wrong, run that through sovitsSVC and than run the final combination of the vocals thought a 10% autotune to fix all the strange pitch ups/downs that were impossible to remove (there is a reason why I am not volunteering to voice over stuff). For some reason none of the ai conversion tools were able to create the word "perfect" in an acceptable degree.
Final result, very much OK, however given that I know nothing about writing my own music it's still better than expected.
Lyrics:
Mister nonny
Mister nonny, ya all right?
We haven’t seen you rest of the night
Mister nonny, you are caked in grease!
We don’t really get it, your odd master piece.
The others may not really get it,
but it makes perfect sense to me!
Mister nonny
Mister nonny, it’s so cute! That’s a facts!
We seen you chow down all of the snacks,
cheese and crackers, an ‘shroomy omelette
and even fishies from Flutters too~
And yet… Mister nonny
Mister nonny has some hoom~man cravings~~~
For there are things that ponies can’t do…
and that very very odd thing will be~
The! Slice! Of! Moooooooon cheese!
[bridge]
Mister Nonny - Mister nonny it’s the truth!
it’s a brilliant master plan
to gotcha yourself a piece of space snack.
Your determination, it shines so bright, Reaching for the stars, taking flight.
Though others may not understand,
I'll stand by you, hoof in hand.
And though it may seem rather queer,
Your fantasies, to me, are oh so clear.
Mister nonny,
Let's not delay, Your rocket ship will lead the way.
[slower outro]
OG video uploaded by BGM : https://www.youtube.com/watch?v=PV2P7HwrS5g
My apologize for missing first 15 seconds of the footage, the OBS failed on me for no reason and I had to restart it.
So here we are, with 5th redub of the MLP from PPP.
Parody of the song by classic Blue Oyster Cult - (Don’t Fear) The Reaper.
Lyrics:
https://files.catbox.moe/2vz7o3.ods
Music:
https://files.catbox.moe/4rlbnj.mp3
https://files.catbox.moe/qtwodh.wav
Made by first creating lyrics with ChatGPT, than converting them to singing voices with Synth V, than use audio clips of that on Talknet to have more naturally sounding voices than use those with the so-vits-svc.
Art made with the Stable Diffusion Pony Soup model (and a lot of img2img in-painting and ms paint correcting)
Timestamps:
Introduction 00:00
Step 0a Gothic Sourcer 00:25
Step 0b Gothic VDFS 02:44
Step 0c Github Python files 04:11
Step 1 separate dialogue, audiofile names and rest of the code script 04:50
Step 1a Polish letters fix 05:23
Step 1b divide NPC and Player dialogue 05:44
remove _output text files 06:15
Marge player character lines 06:30
make temporary audio file lists 07:48
Step 7a move wav audios from speech folder 08:10
Step 7b remove the temporary audiofiles 08:39
rename the files 08:55
Download list:
Python files:
https://github.com/AmoArt/Gothic-Dialogue-Process-dataset
Gothic VDFS (this one should come automatically with Gothic2 installation tools folder):
https://worldofplayers.ru/threads/42314/
GothicSourcerV3_14:
https://worldofplayers.ru/threads/38318/
################################################
NEW code update.
IF you are working with German or English version of the game, run the code below:
Run "_dialogue_1_seperator_005a" in the "Dialoge" folder, to generate "output_*.txt" files.
IF you are working with Polish version of the game, run this (ignore instructions regarding the "_dialogue_1b_letters_convert_POL", that code is outdated.):
Run "_dialogue_1_seperator_005b_POL"
IF you are working with Russian version of the game, run this:
Run "_dialogue_1_seperator_005b_RUS"
Start 00:00
Create conda env 00:18
Testing CUDA 17:28
Connecting Colab ipynb to GPU 18:23
Start of offline code 21:59
Inputting audio dataset 28:53
Final training cell run 33:26
Instructions:
https://github.com/AmoArt/UI_for_So-Vits-SVC/blob/main/folderDumpText/offlinesovit/info.txt
File to place in GDrive:
https://github.com/AmoArt/UI_for_So-Vits-SVC/blob/main/folderDumpText/offlinesovit/sovits_EN_training_colab_4_0_offline_(231703).ipynb
Warning, this is NOT a proper tutorial, just merely my way of training offline models for So-Vits-SVC 4.0, so this method may work on your pc or may not, due to mysterious ways of google colab and python constant breaking compatibility of modules so I would be prepared to not be surprised if this code do not work on your machine.
Timestamps:
00:00 - preparations
00:18 - Step 1 Gathering your audio data
01:41 - Step 2 Setting Up Your Colab Script
02:50 - Step 3 Training Your Model
17:32 - Step 4 Synthesizing Your Audio
Links:
PPP quick guide
https://docs.google.com/document/d/1PDkSrKKiHzzpUTKzBldZeKngvjeBUjyTtGCOv2GWwa0
PPP Main Guide
https://docs.google.com/document/d/1y1pfS0LCrwbbvxdn3ZksH25BKaf0LaO13uYppxIQnac
(So-Vits-SVC 4.0 training Colab code)
https://colab.research.google.com/drive/1laRNiMSgSw_SxSnuti8oWIuC--RHzAGp
(So-Vits-SVC 4.0 synthesis github code (3.0 voice models will not be compatible))
https://github.com/effusiveperiscope/so-vits-svc/tree/eff-4.0
-------------------------------------------------------------------
(this is old v3.0 Colab that is presented in this video, I recommend using the above newer 4.0)
https://colab.research.google.com/drive/1rCUOOVG7-XQlVZuWRAj5IpGrMM8t07pE?usp=sharing#scrollTo=-hEFFTCfZf57
-------------------------------------------------------------------
Video Transcript:
DESCRIPTION COPY:
https://github.com/AmoArt/UI_for_So-Vits-SVC/blob/main/description_YT_links
-So-Vits-SVC 4.0 synthesis github code (3.0 voice models will not be compatible)
https://github.com/effusiveperiscope/so-vits-svc/tree/eff-4.0
-So-Vits-SVC 3.0 : https://github.com/effusiveperiscope/so-vits-svc/tree/eff
Colab so-vits-svc 4.0 (set up to use the Vul PPP models)
https://colab.research.google.com/drive/1128nhe0empM7u4uo5hbZx5lqjgjG1OSf
Colab so-vits-svc 4.0 (set up to use the custom zipped models)
https://colab.research.google.com/drive/1xUK-bdsrGoV5PUJz3z1cArTH0of1CWON
//////////////
-Models (3.0 and 4.0):
https://docs.google.com/document/d/1y1pfS0LCrwbbvxdn3ZksH25BKaf0LaO13uYppxIQnac/edit#
-hubert model:
https://github.com/bshall/hubert/releases/download/v0.1/hubert-soft-0d54a1f4.pt
-rubberband library:
https://breakfastquay.com/rubberband/
###############################
cd (dir)
conda create --name so_svc2
conda activate so_svc2
pip install -r requirements.txt --extra-index-url https://download.pytorch.org/whl/cu113
pip install setuptools==57.5.0
pip install praat-parselmouth librosa tensorboard torchvision==0.11.1 --extra-index-url https://download.pytorch.org/whl/cu113
pip install PyQt5
--- https://breakfastquay.com/rubberband/
--- Environment Variables
;C:\rubberband-3.1.2-gpl-executable-windows
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
pip install pyrubberband
pip show pysoundfile
pip uninstall pysoundfile
^^^^^^^^^^^^^^^^^^^^^^^^^
pip install soundfile==0.10.3.post1 --force-reinstall
pip install requests
python inference_gui2.py
======
-- Get models voice model into 'models' folder
-- Get hubert model into 'hubert' folder
https://github.com/bshall/hubert/releases/download/v0.1/hubert-soft-0d54a1f4.pt
hubert-soft-0d54a1f4.pt
======
python inference_gui2.py
#####################
music used in video:
Seven Songs for Seventh Saga - III. Star
Mela_-_08_-_Pegasus
Sergey_Cheremisinov_-_01_-_Black_Swan
sawsquarenoise - fantasia de los pinos - 05_-_Rito_Oculto
Happy, Train, Generic background ~ FiM_s02e14-2
From the Viva Frei channel:
"Feb. 19 - Live from Ottawa - Viva on the Street"
https://www.youtube.com/watch?v=Zeg8QV-tiOE
SOURCE:
https://www.youtube.com/watch?v=_LAaoxa14L8
LOTR GoT Woke | Hollywood vs Gina | MSM vs Joe Rogan - Friday Night Tights #183 w/ Nick Rekieta
Cover made using the Twilight Talknet model. I would like to thank Clipper for help with getting the reverb at correct pace.
Animation done in blender, you can see the whole process of making the video here:
https://www.youtube.com/playlist?list=PLGvMe2AuzGjk870YnDlyRaLEu_KwOYwdd
Link to mp3 file:
https://u.smutty.horse/mfnatamfdgy.mp3
Source "RazörFist Arcade: MECHWARRIOR 3" :
https://www.youtube.com/watch?v=wwUxhl1tmm8
MP3: https://u.smutty.horse/mefaukbvgpb.mp3
Lyrics are from a 2015 thread were anons wrote the parody of the Hellfire from Disney's Hunchback of Notre Dame.
Song was made using Talknet Fluttershy model with reference audio sang by yours truly.
To celebrate 100th thread of continued posting and advancement in the AI voice synthesis PPP Anons decided to re-dub the first episode of the MLP Friendship is Magic into one episode long YTP.
For anybody whom missed the episode, here is the recording of the 100th thread celebration abridged episode made by the PPP anons with with the rewatch chat, enjoy folks.
00:00:00 EP1
00:26:14 EP2
00:51:12 1-1 rejected clip
01:03:12 Unused Scenes
Source: "Reject Modernity. Embrace Pulp Heroism."
https://www.youtube.com/watch?v=2F6qWGob5tE
disclaimer: im not a doctor, just your average online meme maker that is just passing the information to the rest of the world.
What you do with this is none of my concern.
Original review document:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8383101/
Archive back up:
https://web.archive.org/web/20210906142115/https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8383101/
music used:
00:00 Yakov_Golman_-_01_-_Air.mp3
02:22 Yakov_Golman_-_02_-_Breeze
04:32 Yakov_Golman_-_03_-_Hope.mp3
06:32 Yakov_Golman_-_03_-_Infinity
09:54 Yakov_Golman_-_05_-_Japan.mp3
12:35 Yakov_Golman_-_06_-_Peaceful.mp3
14:59 Yakov_Golman_-_07_-_Reflection.mp3
and if someone wants to hear an actual expert opinion, this man made a pretty good video on a different study related to this thing:
Dr. John Campbell, "Best ivermectin meta analysis"
https://www.youtube.com/watch?v=3j7am9kjMrk
Original story written by MythrilMoth:
https://www.fimfiction.net/story/310636/fluttershys-extremely-odd-jobs
If you wish to contact me or follow me on social media sites, links are here:
https://allmylinks.com/amoart
Using lattes AI tech to have computer generated voice of Twilight Sparkle singing the Adventure Time credit song.
source: NTD "LIVE: Senate holds 1st federal hearing on election 'irregularities' (Dec. 16) | NTD"
Another Video Removed by Youtube's 'Ministry of Truth'
Original story written by MrEnter: https://www.fimfiction.net/story/108236/clean-up-on-smile-five
If you wish to contact me or follow me on social media sites, links are here:
https://allmylinks.com/amoart
Texas Senate Committee on State Affairs, May 6th, 2021.
This hearing starts at 44:24 .
https://tlcsenate.granicus.com/MediaPlayer.php?clip_id=15926
Music used:
00:00:00 Sea by Portrayal https://freemusicarchive.org/music/Portrayal
00:03:56 Just Beginning by Ryan Sheeler https://freemusicarchive.org/music/Ryan_Sheeler
00:07:31 Check Them In by Ema Grace https://freemusicarchive.org/music/Ema_Grace
00:11:20 Return (unknown artist)
00:12:08 Dufferin County by Drake Stafford https://freemusicarchive.org/music/Drake_Stafford
00:15:30 Suco de Abacaxi by Guifrog https://freemusicarchive.org/music/Guifrog
00:18:48 Small Town by Drake Stafford
00:22:07 Humming for you by Ema Grace00:26:43 Find Out When We Get There by Derek Clegg
00:29:28 Young Lovers by Mise Darling https://freemusicarchive.org/music/Mise/Rebel_1433/Mise_Darling_-_Rebel_-_14_Young_Lovers_1196
00:34:01 Pegasus by Mela https://www.freemusicarchive.org/music/Mela00:36:37 In My RPG by Javolenus
00:40:24 Superoro Quin Conoce el Camino by Silva de Alegria https://freemusicarchive.org/music/Silva_de_Alegria/Primavera_en_la_Guerra_del_Sonido/Silva_de_Alegra_-_Primavera_en_la_Guerra_del_Sonido_-_07_Superoro_-_Quin_Conoce_el_Camino-
Blender 2.79b :
https://download.blender.org/release/Blender2.79/
Gothic Blender Files "New World" and "Old World" :
https://github.com/AmoArt/Gothic-Files-Storage/tree/main/Blender
"How do I extract textures out of the Gothic 2?"
All is explained in here :
https://www.bitchute.com/video/lAZ3OsFlQ40e/
BTW, this isn't tutorial on how to use and model in Blender.
I've saw this discussion in the morning and couldn't resist myself to make it into a video shitpost.