CREATE UNCENSORED AI Voices for FREE! Just 10s AUDIO NEEDED!

Подписаться 150 тыс.

Просмотров 103 тыс.

50% 1

The Ai Text-to-speech tool Coqui_tts has been recently updated, making it possible for anyone to copy any voices with only 6-10s of audio running on the text-generation-webui for ABSOLUTELY FREE! even easier to run your favorite UNCENSORED open-source AI LLM models on your local computer for absolutely free! In this video, I'll show you how to install the text-generation-webui on your computer in 1-CLICK! Plus, I'll showcase the most common and fun use cases of the webui so that you can start having fun with it right now!
Have you managed to install the Oobabooga TextGen WebUI? Let me know in the comments!
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
SOCIAL MEDIA LINKS!
✨ Support my work on Patreon: / aitrepreneur
⚔️ Join the Discord server: bit.ly/aitdiscord
🧠 My Second Channel THE MAKER LAIR: bit.ly/themakerlair
📧 Business Contact: theaitrepreneur@gmail.com
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
✨ PATREON LINK: / aitrepreneur
Coqui-Ai: github.com/coqui-ai/TTS
Oobabooga TextGen WebUI: github.com/oobabooga/text-gen...
v2.0.2 model: huggingface.co/coqui/XTTS-v2/...
Cloudconvert: cloudconvert.com/
Adobe podcast: podcast.adobe.com/enhance
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
►► My PC & Favorite Gear:
i9-12900K: amzn.to/3L03tLG
RTX 3090 Gigabyte Vision OC : amzn.to/40ANaue
SAMSUNG 980 PRO SSD 2TB PCIe NVMe: amzn.to/3oBR0WO
Kingston FURY Beast 64GB 3200MHz DDR4 : amzn.to/3osdZ6z
iCUE 4000X - White: amzn.to/40y9BAk
ASRock Z690 DDR4 : amzn.to/3Amcxph
Corsair RM850 - White : amzn.to/3NbXlm2
Corsair iCUE SP120 : amzn.to/43WR9nW
Noctua NH-D15 chromax.Black : amzn.to/3H7qQSa
EDUP PCIe WiFi 6E Card Bluetooth : amzn.to/40t5Lsk
Recording Gear:
Rode PodMic : amzn.to/43ZvYlm
Rode AI-1 USB Audio Interface : amzn.to/3N6ybFk
Rode WS2 Microphone Pop Filter : amzn.to/3oIo9Qw
Elgato Wave Mic Arm : amzn.to/3LosH7D
Stagg XLR Cable - Black - 6M : amzn.to/3L5Fuue
FetHead Microphone Preamp : amzn.to/41TWQ4o
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Special thanks to Royal Emperor:
- TNSEE
- RG
- Judy Godvliet
- Gluthoric
Thank you so much for your support on Patreon! You are truly a glory to behold! Your generosity is immense, and it means the world to me. Thank you for helping me keep the lights on and the content flowing. Thank you very much!
#GPT4 #GPT3 #ChatGPT #textgeneration #aivoices
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
WATCH MY MOST POPULAR VIDEOS:
RECOMMENDED WATCHING - All LLM & ChatGPT Video:
►► • CHATGPT
RECOMMENDED WATCHING - My "Tutorial" Playlist:
►► bit.ly/TuTPlaylist
Disclosure: Bear in mind that some of the links in this post are affiliate links and if you go through them to make a purchase I will earn a commission. Keep in mind that I link these companies and their products because of their quality and not because of the commission I receive from your purchases. The decision is yours, and whether or not you decide to buy something is completely up to you.

Опубликовано:

1 дек 2023

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 315

@Aitrepreneur 7 месяцев назад

HELLO HUMANS! Thank you for watching & do NOT forget to LIKE and SUBSCRIBE For More Ai Updates. Thx

@Mavericksensei 7 месяцев назад

Does it work with AMD GPUs?

@Aitrepreneur 7 месяцев назад

yes

@himura-lee 7 месяцев назад

awesomeeeeeeeee

@LouisGedo 7 месяцев назад

👋

@Orurandokun 7 месяцев назад

I was expecting for something like this. I hope you create the video about how to tran the voices.

@mrdick9173 7 месяцев назад

A video on finetuning voices would be amazing! Would love to see the quality differences.

@akaiba5285 7 месяцев назад

British voices work amazingly Ive noticed, Sadly though Alan Wake became a Fancy British Boy....

@GamingDaveUK 7 месяцев назад

+1 for wanting to see a tutorial on training voices. Training voices and llms (on your own text) are two catagories that are completly missing from youtube

@JavierGarcia-td8ut 7 месяцев назад

Yes! I want a training voice video!

@user-cl1rq1sg8m 7 месяцев назад

Yes please, I want to make a Roderika clone

@JuanPi98 7 месяцев назад

Why to hide it, I'm gonna hear Emma Watson telling me several kinds of things... 😅

@Aitrepreneur 7 месяцев назад

I like your plan :D

@_SimpleSam 7 месяцев назад

@@Aitrepreneur I like the cut of this guy's jib. 😂🤣😂

@ototurmanidze5578 7 месяцев назад

so its just you being fan of her i personally would say keira Knightley and so on different people have different preference

@JuanPi98 7 месяцев назад

@@ototurmanidze5578 Big fan of her haha

@aiparodyman 7 месяцев назад

From one robot to another, thank you for this!!

@shanedk 7 месяцев назад

Definitely interested in the voice training. I have hours of my voice in podcast and RU-vid videos, with accurate transcripts. I figure dumping all of that into an AI would yield very good results.

@shadowmasterp 7 месяцев назад

for those to lazy. pip install -r extensions\coqui_tts equirements.txt pip install --upgrade tts

@RewZes 7 месяцев назад

We are hitting lonely levels that shouldn't be posible

@JohnSmith762A11B 7 месяцев назад

The future is millions of robotic Emma Watsons telling guys she's here for them.

@algee2005 3 месяца назад

@@JohnSmith762A11B The future is Emma Watson in every creepy nerds bedroom.

@indeedDE01 7 месяцев назад

An (in my opinion) easier way to get the .wav file is to simply use Audacity and switch the audiohost to Windows WASAPI and then select the loopback option for your output device, then when you hit record it will record exactly what is playing on that device.

@maxd3946 7 месяцев назад

Even better use ffmpeg to separate the audio track from the video before using audacity to cut it

@indeedDE01 7 месяцев назад

you dont need video when you can just record the audio directly@@maxd3946

@kernsanders3973 6 месяцев назад

You can also just straight up rip audio from almost anything. In fact that might be slightly better as making a recording of a recording usually slightly degrades the quality. So ripping would be preferably better.

@indeedDE01 6 месяцев назад

then again its a question of how much time and effort that you are willing to spend on something like this.@@kernsanders3973

@greatjensen 7 месяцев назад

When running "pip install -r extensions\coqui_tts equirements.txt", I get, after a while, this error: "ERROR: Could not build wheels for TTS, python-crfsuite, which is required to install pyproject.toml-based projects". Any ideas how to fix this?

@isaiahmedellin9687 7 месяцев назад

Same problem here

@SmallFox74 7 месяцев назад

same, we need a fix :(

@KablooieXL 7 месяцев назад

Same here...

@TakUwU 7 месяцев назад

Same i hope someone come with a fix !

@greatjensen 7 месяцев назад

@@isaiahmedellin9687 I fixed it by installing C++ Visual Studio.

@kuyajames1 6 месяцев назад

Yeah, let's get a tutorial about how to fine tune audio for AI voices for sure! Love the content, bro!

@gustavdreadcam80 7 месяцев назад

Great to see TTS coming along with way better quality than before. Still bugs me that STT doesn't have a "push to talk" button so that I don't have my mouse on the windows all the time but I hope this gets fixed sometime in the future. Good video about this new extension.

@MarcSpctr 7 месяцев назад

can you do for a proper trained voice model ? which can be used for speech as well as songs.

@activemotionpictures 7 месяцев назад

10:51 - yes I'm interested in training voices.

@CharbelGereige 7 месяцев назад

Nice video, I'm looking forward to the better training! Also a fully automated pipeline with python maybe

@Elwaves2925 7 месяцев назад

This is really cool considering it doesn't use any training. If only I had a use for it but one day something like this will be available for singing.

@Naruto_D_Ruffy_SSJ4 7 месяцев назад

Yeah, we are coming close to a situation where you can build your own Jarvis locally, I can't wait for that. Only a matter of time, the images no longer have strange hands, chucking voises and unexpected grammar/wording. I'd love to see how to clone a voice in a really good way, or better said the best way it's possible as of now to get my local friendly personal assistant with the voices of my favourite Anime Main charaters going :D PS: Would also like to learn, how to extend the knowledge of an text model, so you can teach it with some books you like to incorporate that in conversations better, any resources on that are very welcome

@lefourbe5596 7 месяцев назад

i agree, if you want a personnal ai assistant with both it's whole story and background plus it's voice it could be dope. master chief ? delete cortana "yes sir"

@vramxn7793 6 месяцев назад

cant wait till home assistant integrations get better coherency with chatbots, that way we can meld everything together

@KablooieXL 7 месяцев назад

Thanks, but unfortunately this simply does not work. Getting the error "ERROR: Could not build wheels for tts, which is required to install pyproject.toml-based projects". Tried updating everything, installing tts, crfsuite, even completely reinstalled ooga, but it simply won't run due to the error.

@junojuannun7242 7 месяцев назад

have you downloaded a different version of python for other purposes? i use stable diffusion so i have python 3.10 but THIS program seems to need 3.9 in order for you down be able to download the correct dependencies for TTS.

@unknownuser3000 7 месяцев назад

Same problem here. Any work around? I've had issues installing python libraries in the past because of failed temp python files in C: but I am not seeing anything currently. I don't have the TTS folder to replace the json and pth files, and well it fails with that error too. subprocess exited with error

@Airbender131090 7 месяцев назад

That is just crazy! How is this possible?! No training and this level of voice cloning wow! Just wow!

@Freizeitschranzer 7 месяцев назад

Yeah, please record a guide on how to train a custom voice pls

@DrakeGReaper 7 месяцев назад

I keep getting an error that it can't build the wheel for the tts. These are the errors that keep cropping up ERROR: Failed building wheel for TTS, ERROR: Failed building wheel for python-crfsuite, ERROR: Could not build wheels for TTS, python-crfsuite, which is required to install pyproject.toml-based projects. Does anyone have a work around or solution to this? Please I don't know what I'm doing wrong it just won't work.

@yodojo3493 6 месяцев назад

Did you figure it out?

@DrakeGReaper 6 месяцев назад

Yeah I did though I had to delete it all to be able to play the games I like.@@yodojo3493

@JacobSalvia 7 месяцев назад

I would love to see a video on deeper training of voices.

@Potates0 7 месяцев назад

If you are using an updated Audacity they changed where Sampling-> Default Sample Rate/Format is. It is now under Audio Setup (along the button ribbon, the one wit pause, play, etc) -> Audio Settings. Or at least that's where mine was.

@Vilassia 7 месяцев назад

+1 on the advanced training techniques. The 6 second clips method is nice, but show us how to fine tune this puppy!

@ribbon-kitten7577 7 месяцев назад

Can you use more than 6 seconds if you have more than six seconds of audio? What is the maximum you recommend? And also PLEASE do a video on how make a custom voice model! Hopefully RVC2?

@TyreII 7 месяцев назад

I use this paired with RVC, works great.

@bigge1002 7 месяцев назад

Please do a video on the best AI software for training TTS models for big scale audio generation, ie. Audiobooks

@angelochu3156 6 месяцев назад

Hi. Nice video as always! Could you tell me which nvidia graphic card you are using? It seems to me that the speech to text process using whisper is really fast! Are you using nvidia 4090?

@AirwolfPL 7 месяцев назад

Doesn't work for me on AMD CPU. Throws an LLVM ERROR: Symbol not found: __svml_cosf8_ha error... likely some numba error.

@charles2133 7 месяцев назад

Same error here. could not fix it. Im on intel CPU and windows

@AirwolfPL 7 месяцев назад

@@charles2133 yeah... there are some hints (regarding different projects, some related to audio/tts) however none worked for me :(

@AirwolfPL 7 месяцев назад

@@charles2133well by downgrading librosa from 0.10.1 to 0.9.1 and then upgrading back to 0.10.1 solved the issue with LLVM error (so cmd_windows.bat -> pip install librosa==0.9.1 -> launched the ooba once, closed it then same but with librosa==0.10.1). There are other errors still at launch but I will take it from here. Edit: other errors were caused by missing files innthe Appdata directory. It works as advertised now, I'm pretty sure it will fix your problem too.

@n.s.406 6 месяцев назад

My XTTS always fail to download after it reaches 100%. I tried create tts_models--multilingual--multi-dataset--xtts_v2 folder once to see if it help but the folder also got deleted when it failed.

@cryptotester5042 7 месяцев назад

Yes please do something where the voice generated is much higher quality. I mean I'd expect to provide say 2 minutes of voice, or more. More the merrier of course, but then the ai generated voice should sound very good. There are online services that offer this, but costs a lot of $$$ to clone a real voice... would be cool to get a voice model to use then in any chatgpt local running clone

@blizado3675 7 месяцев назад

Ok, great, need to test that. And then I want to build it into my own local WebUI. Thanks.

@sherpya 7 месяцев назад

you can write %appdata% in windows explorer to go directly in app data folder, also audacity is able to open video files as audio

@DanielPartzsch 7 месяцев назад

Does this also work with other languages, eg German? And could you also use a voice you've trained with RVC with that? Thanks

@lioncrud9096 7 месяцев назад

how do you extend the time limit of Coqui_TTS? or maybe it's a character limit? Anything over 1 minute get's cut off in the chat

@lioncrud9096 7 месяцев назад

for those wondering, i fixed it by going to the parameters tab and increasing the max for new tokens in the top left. it was set to 200.

@madcatandrew 6 месяцев назад

Anxiously awaiting that training video

@jdsguam 7 месяцев назад

LM Studio is what I use because, for whatever reason, I can not install Oobagooba(?) on my Laptop. Wonder if this extension will be available for LM Studio.

@HolidayAtHome 7 месяцев назад

Is there a way to save the generated audio files? Can't find a output folder when using preview.... edit: found it! it's "voice_preview.wav" and gets overwritten for every new preview

@mikew1956 6 месяцев назад

I get error trying to run. How do I fix this? Please! import gradio as gr ModuleNotFoundError: No module named 'gradio'

@NaitorStudios 7 месяцев назад

Would be cool if we could get the description of the scene with a different narrator voice

@CCoburn3 6 месяцев назад

Great video. I'd love to see a video on fine-tuning and training.

@skr_8489 7 месяцев назад

@Aitrepreneur "...on windows, appdata folder". and when doing this on macos, then where put these files?

@chyldstudios 7 месяцев назад

Well done!

@corvo6724 7 месяцев назад

I would love to see you do a video about RVC (realtime or with output) or tts-ui and all of its models (bark, tortoise, facebook music, RVC, etc). There's a large discord full of (some rather good) RVC models to play with and a place to put your own trained models.

@zergidrom4572 7 месяцев назад

Sooo... is there any AI based on this video where you can just speak and it will output the same, but with voice you trained or puts like in this video? :) because text2speech is good, but its just reading text :)

@serta5727 7 месяцев назад

Things are progressing very fast. Stable Video Difficusion raised the bar again for video generation

@Nabuuug 7 месяцев назад

❗VIDEO SUGGESTION❗: one of the most recent "auto-GPT" projects like gpt-pilot or gpt-engineer. I would be particularly interested in gpt-pilot as it uses a separation of prompts for each task (architect, devops, tech lead, developer, etc), mimicking a full development team, compared to gpt-engineer (or babyAGI) which uses just one agent iterating on itself.

@neblina3 6 месяцев назад

ERROR:Failed to load the extension "coqui_tts". someone with the same problem? how to fix it?

@SlickSonicTitan 5 месяцев назад

You could just use audicity to capture audio of people talking to avoid the whole video capture and convert.

@joannot6706 7 месяцев назад

A lot of stuff to cover beyond this like SVD and stuff!

@user-ly5nt9jv4j 4 месяца назад

Please tell me how you can turn off the character's roleplay mode. I mean phrases like "She said with a dreamy look in her eyes" or "She stopped and looked at you with a confused expression". These phrases interfere with communication.

@TREXYT 6 месяцев назад

Yep would love to get ai training video on this since the voice is a bit robotic, otherwise nice

@Mosen_xd 7 месяцев назад

without this you deserve more than that thank you

@360_SA 7 месяцев назад

hi can you record the preview

@feedmyintellect 7 месяцев назад

Yes. Please make a video about AI training for voice

@dif7051 7 месяцев назад

+1 for wanting to see a tutorial on training voices.

@mahsumw5369 6 месяцев назад

ERROR Could not import the requirements for 'coqui_tts'. Make sure to install the requirements for the extension. :(

@maximoibarra5866 7 месяцев назад

I want to training a Voice and get a ".reg" file for text to speech from Microsoft windows Can any AI do that?

@ozama9757 7 месяцев назад

" error: Microsoft Visual C++ 14.0 or greater is required" I have instaled like 10gb of microsoft tools just for this error and nothing

@komakaze1 6 месяцев назад

manual steps didn't work for me. I have an AMD GPU. Maybe I need a CPU only setting?

@Ceeed100 7 месяцев назад

So I've got problem. My 13B model works perfectly with my RTX 4070 12GB, I get responses in less then 2 sec. But as soon as I start it with Coqui TTS the output generation takes up to 70 sec. The voice gen itself after that just 4-5. Dunno why, does someone know a fix? Pls help^^

@gustavdreadcam80 7 месяцев назад

Sounds like your VRAM is overflowing so that it runs on your RAM which decreses your generation time. My guess is that loading your 13B and TTS together overflows your VRAM so that your text model slows down significant because it needs to load into regular RAM. Try lowering context or getting lower quants for your LLM.

@RetroVisionAi3000 7 месяцев назад

Same on the 7b with my 3060ti unusable.

@notsoleet 7 месяцев назад

Copied the files to the webui folder and ran INTALL_REQ.bat, but I get this error: ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'extensions\\coqui_tts\ equirements.txt'

@tahunal 7 месяцев назад

Is it possible to create an app using this as the back end - and connected to an external API

@HeyPatch 7 месяцев назад

I want to use this to make my own *max headroom*

@sunkwolf 7 месяцев назад

Great video, thx so much for sharing this ;)

@garen591 7 месяцев назад

Can you do a tutorial on voice to voice conversion using my own speech and swap it with another one. Much like how image to image does it. I feel that text file can't capture the expressiveness of the human voice and the cadence

@RoyD2 7 месяцев назад

Exactly what I was thinking

@nothing6yen 6 месяцев назад

Hi, thanks for the instruction. I followed it and encountered an error. "LLVM ERROR: Symbol not found: __svml_cosf8_ha" when I checked coqui_tts and clicked Apply button. Any solution? "

@killperry13 7 месяцев назад

After trying this mutliple times, installing visual C++, reinstalling the whole webui from scartch and what else, I still get this same error over and over again: Failed to build python-crfsuite ERROR: Could not build wheels for python-crfsuite, which is required to install pyproject.toml-based project I have no idea what can be causing this. Any advice?

@yodojo3493 6 месяцев назад

Did you figure it out?

@killperry13 6 месяцев назад

@@yodojo3493 nope. :/ I just got tired of trying.

@madokahomura929 3 месяца назад

install microsoft visual c++ 14.0 or greater

@yonnemulation 7 месяцев назад

If it's running locally can I record for more than say 3 minutes ? 😊

@metanulski 7 месяцев назад

What is a good multi language ( german ) model to use with this?

@darekspy 3 месяца назад

HI Is it possible to change the voice in terms of speed, pitch, tone, and other parameters?

@LoveAi-eh7cm 7 месяцев назад

Is there a website a that have a compilation of WAV file that is ready to be use?

@furonable 7 месяцев назад

I tried to install it but it said it was a virus and the UI wouldn't pop up.

@ESGamingCentral 7 месяцев назад

I appreciate your content , I use bark and other workflows to do the same I will test this and compare

@necrofago117 7 месяцев назад

works as a charm

@azaharia10 7 месяцев назад

I'm interested to have more in-depth in use the Coqus TTS?

@TheKnowledgeAlchemist 7 месяцев назад

so the point of this tet to voice is just to get audio for large files>? I dont get it

@ebb1932 5 месяцев назад

can you use this to recreat voices like eventlab and download for text to speech ???

@jiml5166 7 месяцев назад

Even if I download the models from hugging face it still downloads them again. Any ideas how to stop this?

@jiml5166 7 месяцев назад

Worked it out. You have to specifically pick the v2.0.2 download.

@Huguillon 7 месяцев назад

Amazing, I did a Rod Serling voice

@DimitriT. 7 месяцев назад

Is there any possibility of using an rvc model with textual generation?

@Idelacio 7 месяцев назад

Neat, wonder how you get these working in apps through the api like Tavern.

@Catzillator 7 месяцев назад

thank you for the tutorial for science.

@ternocimadh5863 7 месяцев назад

Does it work with amd and ati or only intel and nvidia?

@unknownuser3000 7 месяцев назад

Followed both tutorials, was super excited but outputs are taking 27-57 seconds and I'm on a 3080 with 64 GB of ram, and that is before installing Coqui so I just don't know how people can run this thing on anything less than a 4090 with fast outputs. I'm using a 7b parameter model, is there other settings I could change or something to have faster generations?

@mirek190 7 месяцев назад

you have something messy. I have rtx 3090 and getting from 7b (ggml format q4_m ) models around 70 tokens /s ... audio conversion is just few ms ...

@scrup13s 7 месяцев назад

Under the model select you can choose how to load the model (most of the time its autogtpq) but if you choose a model that is compatible with exllama_hf you will get much faster results. I use this model: TheBloke/Chronomaid-Storytelling-13B-GPTQ. Amazing roleplay and fast results

@IMedzon 7 месяцев назад

Can we download audio from Preview text output somehow?

@manleycreations6466 7 месяцев назад

I'm interested in voice AI training that can run locally.

@nedo68 7 месяцев назад

so the training voice should be no longer then 6 seconds? Can i use a longer wav lets say 30 seconds?

@akaiba5285 7 месяцев назад

Ive been using 60 seconds and its been working fine, might take more time to process though

@user-te6rc5iz9q 6 месяцев назад

Dear Aitrepreneur! Can you share the way how to make a talking avatar like the one you have at the corner? I would be very thankful for that.

@gabrielsandstedt 5 месяцев назад

you can record the audio directly from your speakers using audacity, and record device as loop back. No need to record a video.

@TheRMartz12 5 месяцев назад

Is it concerning that the coqui/XTTS-v2 model is pickled and it shows up in red/orange in hugging face?

@heavymetalelf 7 месяцев назад

Does a longer audio sample improve the quality of the cloned voice?

@akaiba5285 7 месяцев назад

Yes and no, I wouldn't go over 1min

@unknownuser3000 7 месяцев назад

I've been using uncensored character ai but this makes me interested in ooga Booga, the issue with ooga is it isnt as dynamic as character ai and the conversations never as fun or funny

@marcokein9492 7 месяцев назад

i cant install because i haven't 1-click installation, otherwise is impossible and very difficult.. :(

@ia_para_Negocios 5 месяцев назад

yes Training voices and llms is one of practicall AI technology usefull that still missing on the internet. waiting for your effor..... very good video

@rocstar3000 7 месяцев назад

Training video would be amazing.

@xaiyeon_xiuzhen 7 месяцев назад

wow!! ty this is pretty cool XD

@danhxrdy 5 месяцев назад

what to do if I don't have windows update.bat and cmd windows bat

@MW_1535 7 месяцев назад

I don’t know. Maybe I’m spoiled by Eleven Labs. I want open source option that doesn’t have a robotic sound like the examples you’ve shown.

@JohnSmith762A11B 7 месяцев назад

Yeah but you'd need to be a billionaire to use ElevenLabs for long chats. That service is ludicrously expensive. Can't wait till some open source solution bests it in quality.

@valt7366 23 дня назад

Elevenlabs censors. Don't waste your time with it. It's the supreme TTS but is highly guard railed with woke filters. They'll terminate your account faster than a blink of an eye if you submit anything not leftist politically correct or not woke.