I wonder if the group is full or did the link expired, because I tried joining but couldn't. I really wanted to join the group 😂 this whole project that you've created and dedicated is so amazing it keeps leaving me in awe of the possibilities it could achieve. Thank you so much!
All this secret knowledge for free 😳 it is literally the first time i saw someone who gave their all to do this and then basically tell everything to everyone for free ❤️ you rock bro 🔥
my dude u deserve at least ten times the subs you have rn, i think if you start posting more shorts and more videos of you using the waifu voice in fps games, then the algorithm would pick your channel up again
I am more than happy that u even said something about my comment and i somewhat expected it XD and yeah I am your permanent subscriber so when u grow to 1 million I can proudly say I was one of the early supporter
Pro, your editing is great. Seamless cuts, just the right amount of information on the screen and little things like where you are on the screen and video chapters. It's great.
I've been looking to start on a weekend project for personal growth and learning as a developer and this has totally motivated me to learn and understand all that is happening in the background. Thanks for the detailed guide and how open you are about the whole thing! Keep it up dude!
I believe cause he's still small, he's able to answer our questions. Otherwise, once he starts blowing up, it'll be too overwhelming for him to interact with us, unless he streams... and somehow got your question, my question or anyone's questions.
Thank you for this! Couldn't even imagine that this much work has went to this. By the way I noticed that you used my cover of EMAymniam at 13:42, can you add it to the description? Thank you.
Oh thank you so much for pointing it out! I love our AOT covers, especially your latest Hange's Theme, so I wanted to at least use one of them this video! I can't believe you would watch this video haha
@@sociallyineptweeb Thank you so much! Also, been waiting for this tutorial ever since your "So I turned my VOICE into an anime WAIFU using ChatGPT and AI..." It was so awesome!
Bless you for making this guide, You are doing Gods work! I put the C# repo on public with your link posted at the very top of the Repo, thank you for allowing me to do this!
@@sociallyineptweeb You'll have to excuse me, I suffer from stupid Fixed it, also I have experience with larger discord servers so I'll set one up if you want
@@sareanaka It's ok! As for the discord server, @slowsloth4486 said that he would set it up so you'll will have to check with him to see who will set it up first
Whoa, thanks so much for doing this. That repo is super useful, especially if you don't have a clue how to setup and route audio properly. I just took it as a base, added my own tts service and replaced your whisper implementation with open ai's whisper API veeery easily.
Damn. The fact that YOU don't have more than a million+ subscribers with SUCH content is worse than the warm side of the pillow. You really deserve a lot of subscribers
Thanks for this elaborated guide, you are making world a cuter place. I think you've done a great job with explaining this for non expert users, but even so there always will be people who will struggle with most primitive tasks...
Lol man I saw the first video and decided to do it myself the funny part is I also decided to do a GUI and ended up using CTk, my GUI is way prettier tough, and have some added functionality like TTS aside from recording, was a fun project, ty for the idea.
Your setup Guide was what i needed thnx a lot But seriously for me its not your programming skills but your thought process that truly astonishes me man Gud job ur Awesome
Suggestion tangential to 17:12, Pygmalion-6b is fantastic as a conversational AI similar to ChatGPT and Character AI, and you can run it completely locally. However it's model is absolutely enormous, far too big for an RTX 3060's VRAM pool. I have to use a disk cache to get it running on my current rig, another impetus towards building a dedicated AI compute machine.
This is such an awesome project for you to have taken on, bravo! I was just wondering if there is a reason WhisperAI was used for translation of subtitles and not a translation software like DeepL? I think that would improve the accuracy somewhat? Anyways this is such great work and you definitely deserve more recognition for it.
Since whisper comes in-built with a translation feature, I decided to use it since we can skip a step of sending another request to deepl which can reduce the delay of subtitles being shown.
imagine if this project goes even further with an UI where u just enter required stuff choose a voice (not only japanese) and it works :D so if ur anxious u can just sound different
this is the power of Open Source app, everyone can contribute their skills & knowledge to perfect this app . help this guy . whoever has much more advanced skill & knowledge.. Together.......they will perfect it .
7:20 I enumerated audio devices in c# before. first build a list of available devices, then whenever you enumerate or display later, display the list. when adding a device, do a for i = loop, or a for each loop, to make sure you don't add the same device to the list again
Do you think there's more room of improvements for the Subtitler like making it use DeepL and enhancing the way it transcribe audio (it cuts way often unless you perfect the config for the timeouts)? It's a great way to understand Japanese and it could be used variously in different media such as animes, japanese podcasts, games, and more.
Yup there's definitely ways to improve it, I think the most glaring issue is that whisper is unable to handle simultaneous requests at once asynchronously, causing it to drop some requests.
Any plans to train your own ai? To possibly create a project similar to Neuro-sama, an ai vtuber created by a British dude to play osu, but recently started streaming chating with viewers and sometimes interact with her own creator.
hey man love your videos.. i was wondering... did you try incorporating the NLTK module in python ? to convert English to Japanese so that people who have some low end systems can still run the program when they have no access to the cloud or something just my suggestion as i use the NLTK module a lot.
@@sociallyineptweeb So im just a beginner in python😅.. i don't know as much as you lol😂. But in my experience of using NLTK(natural language toolkit) the accuracy of the sentence may not be as good as compared to the setup you've given us (it's really awesome btw!) but its fine for some low end pc's or people who just don't like the cloud for some reason. And since it is a python module in itself , i am assuming that it would run a little bit faster given that we aren't using DeepL(the middle man you mentioned in your first video about this waifu voice changer). anyway im just a dude who knows the basics of python and some app building skills lol, i don't know too much but ig the NLTK wouldn't make a huge difference other that sacrificing accuracy and gaining a lil bit of performance . looking forward to learn more from your videos btw and keep up the good content man!! btw this is my first hearted comment... 😁😁
In the final step, when i use the comand docker compose up -d i get this error: Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy' nvidia-container-cli: initialization error: WSL environment detected but no adapters were found: unknown any idea? Q_Q
What's the easiest way to test whether its working or not? Should I be hearing the waifu voice playback to me? or would it only play to the others in game? @11:54, what IDLE are you using in this part to see this code? Or where would I see that error if it popped up? I followed along pretty well but I feel like I'm missing something. Edit: I've been trying to follow everything through windows command prompt. I think this is the reason I'm getting errors. Not sure where the best place to run this is.
4:25 When I change the output device to voicemeeter aux input, the subtitler works but I lose the sound from the application. I don't really know what I'm doing so if someone could help me out that would be greatly appreciated
Awesome work man, love the attention to detail :) I wondered would there be a way to specify my second graphics card, to handle both of the programs while i use my main one for playing the game?
I hope you can sell this idea or make your own company out of this idea you've made. I want this to be integrated with all the games MMORPG I know... so badly (T v T)
Hey, thanks a lot for the setup guide. Other than some minor issues, which I managed to google or just reading through discord which I managed to fix. I managed to get it working through using google colab. Got a question though, when using it on discord, do you have to switch the input to cable in order to use voicevox? As far as I'm aware, that seems to be the case, but it won't pick up my mic that way. And I would have to switch it back to mic input for others to hear my actual voice. Is this just a limitation that I can't get around?
@@sociallyineptweeb haha I see. Your editing from when you tested this on discord made me think you might have some alternate methods. Maybe if you could route your mic input into the virtual cable perhaps? Though that would mean have its own problems
Hi, I have a problem: When using Colab, I can't access the voicevox site where I can see the ID of the speakers because I either don't understand how or "replacing the url with the one from the colab" does not work :c Aside from that, I love this project
Discord Link: discord.gg/AUh9XDrvqa
Question anyway I can use it without it draining my memory ?
I wonder if the group is full or did the link expired, because I tried joining but couldn't.
I really wanted to join the group 😂
this whole project that you've created and dedicated is so amazing it keeps leaving me in awe of the possibilities it could achieve.
Thank you so much!
Hey, wouldn't it be better to use papago api from naver? because papago, looks and sound much more natural than deepL ^^'
Now i just need to figure out how to have it work with a english Text to speech Programm. Essentially to switch out my voice completely
what you did here is very cool :D.
Keep it up, much love dude
All this secret knowledge for free 😳 it is literally the first time i saw someone who gave their all to do this and then basically tell everything to everyone for free ❤️ you rock bro 🔥
Sadly my pc is potato so i can't do these but sure 🌝 knowlege is everything
@@PhoenixFromUA-cam Yes you can through Google collab all the processing is done via Google collab
@@PhoenixFromUA-cam thats why he made a vid all about running it on the cloud so you can run it
my dude u deserve at least ten times the subs you have rn, i think if you start posting more shorts and more videos of you using the waifu voice in fps games, then the algorithm would pick your channel up again
He grew 1000% since the ai translator video.
@@bigbirdturkey he went from 2.3k to 15k tf 🥵🥵
mainly needs to start doing better thumbnails
started on my coding journey a few months ago and man does stuff like this get me pumped to keep going
Believe in yourself!
I am more than happy that u even said something about my comment and i somewhat expected it XD and yeah I am your permanent subscriber so when u grow to 1 million I can proudly say I was one of the early supporter
Holy shit I’ve been here since 100 subs and you now have 16k. I still think you’re underrated af. Keep up the good content
Pro, your editing is great.
Seamless cuts, just the right amount of information on the screen and little things like where you are on the screen and video chapters.
It's great.
I've been looking to start on a weekend project for personal growth and learning as a developer and this has totally motivated me to learn and understand all that is happening in the background. Thanks for the detailed guide and how open you are about the whole thing! Keep it up dude!
blah blah blah brain dead noise
You can do it!
@@ahmetselimgt3232 I hope you also fulfill your dreams! I see you commenting positive things and it made me want to say nice things to you too!
@@goldenera777 thanks
thanks for answering my question, you are the real gem man.
I believe cause he's still small, he's able to answer our questions. Otherwise, once he starts blowing up, it'll be too overwhelming for him to interact with us, unless he streams... and somehow got your question, my question or anyone's questions.
@@bigbirdturkey i mean yea thats something
Wooo yeah baby that’s what i’ve been waiting for
Omg let's goooo, you're an absolute king and a great content creator, actually helping the viewers, amazing work
Gotta appreciate you giving yt priority. Tysm ❤️
Thank you for this! Couldn't even imagine that this much work has went to this.
By the way I noticed that you used my cover of EMAymniam at 13:42, can you add it to the description? Thank you.
Oh thank you so much for pointing it out! I love our AOT covers, especially your latest Hange's Theme, so I wanted to at least use one of them this video! I can't believe you would watch this video haha
@@sociallyineptweeb Thank you so much! Also, been waiting for this tutorial ever since your "So I turned my VOICE into an anime WAIFU using ChatGPT and AI..." It was so awesome!
Bless you for making this guide, You are doing Gods work!
I put the C# repo on public with your link posted at the very top of the Repo, thank you for allowing me to do this!
You might have misspelled my name in your readme LOL
@@sociallyineptweeb You'll have to excuse me, I suffer from stupid
Fixed it, also I have experience with larger discord servers so I'll set one up if you want
@@sareanaka It's ok! As for the discord server, @slowsloth4486 said that he would set it up so you'll will have to check with him to see who will set it up first
Aight imma go ahead and sub cause if you keep this up you'll be going places. (inb4 you become mad famous ai programmer)
blah blah blah blah brain dead x 3
Best youtube content for both programming and anime, Loved it man 🔥.
Thank you, your project has brought more waves to my quiet life
Did you even manage to do what he shown here? If not can you shut up
Whoa, thanks so much for doing this.
That repo is super useful, especially if you don't have a clue how to setup and route audio properly.
I just took it as a base, added my own tts service and replaced your whisper implementation with open ai's whisper API veeery easily.
It's amazing job, man. Thank you for that!
Less goo my guy. Thx a lot
Damn. The fact that YOU don't have more than a million+ subscribers with SUCH content is worse than the warm side of the pillow. You really deserve a lot of subscribers
What?
Your comment is kinda contradictory.
You say that he has more than a million subs (he doesn't) and that he should get more subs.
@@arjix8738 probably a bot
@@arjix8738 Yes, I've noticed my mistake, thanks for noticing (and I didn't lol)
@@neonshiji3595 Who knows lol
Thanks.... BRYAN!
Thanks for this elaborated guide, you are making world a cuter place. I think you've done a great job with explaining this for non expert users, but even so there always will be people who will struggle with most primitive tasks...
Lol man I saw the first video and decided to do it myself the funny part is I also decided to do a GUI and ended up using CTk, my GUI is way prettier tough, and have some added functionality like TTS aside from recording, was a fun project, ty for the idea.
Amazing! Hope you improve yourself!
I JUST LOVE THIS OMG IVE DREAMT FOR IT SO LONGGGGG!!!!
Bro deserve 10m subs he's so perfect at programation
"programation"? Are you high
i have no basic knowledge of coding, even no knowledge on python.. i follow your guide.. i did it thank you !
Your setup Guide was what i needed thnx a lot But seriously for me its not your programming skills but your thought process that truly astonishes me man Gud job ur Awesome
Thanks for the repo my man. I will just be using the japanese TTS, but your explanation is very appreciated. I just subscribed.
ayyy free resolve clan
YOUR"RE A LEGEND
ITS FINALLY HERE!
Love your content and constant posts
Suggestion tangential to 17:12, Pygmalion-6b is fantastic as a conversational AI similar to ChatGPT and Character AI, and you can run it completely locally. However it's model is absolutely enormous, far too big for an RTX 3060's VRAM pool. I have to use a disk cache to get it running on my current rig, another impetus towards building a dedicated AI compute machine.
This is such an awesome project for you to have taken on, bravo! I was just wondering if there is a reason WhisperAI was used for translation of subtitles and not a translation software like DeepL? I think that would improve the accuracy somewhat?
Anyways this is such great work and you definitely deserve more recognition for it.
Since whisper comes in-built with a translation feature, I decided to use it since we can skip a step of sending another request to deepl which can reduce the delay of subtitles being shown.
this is first time i watch this content and dayumm i love the youtube algorithm that bring me in this video
did you finish the setup?
imagine if this project goes even further with an UI where u just enter required stuff choose a voice (not only japanese) and it works :D so if ur anxious u can just sound different
Bro is so under-rated.
I just wanted to say hi and thank you for your hard work you’re doing great things!
Hey, today i found your channel and i'm liking it hahahaha. Pd: I'm spanish
Thanks sensei... Now i will exceed my project
this is the power of Open Source app, everyone can contribute their skills & knowledge to perfect this app . help this guy . whoever has much more advanced skill & knowledge..
Together.......they will perfect it .
ur gonna make it big, jesus
OMG i luv you!!!!❤
thank you for sharing your knowledge bro !
do you even get what he did, How do you get it to work?
@@angelwater8987 no bro, i keep getting the docker problem
don't know what's wrong with it
Amazing tutorial vid ty fam!
You’re the best
Did you get this to work? Can you make a tutorial on the usage part?
You have my thanks👍
amazing man!
Finally i find a right way to learn japanese , thank you for sharing your japanese learning progress .
Hope you don't give up unlike me and reach your goal!
oh! Finally....thank you very much
Thanks for shaaaaring
I'm gonna trust this guy and i'm gonna download this because IM A WEEB
DId you get this to work
7:20 I enumerated audio devices in c# before.
first build a list of available devices, then whenever you enumerate or display later, display the list.
when adding a device, do a for i = loop, or a for each loop, to make sure you don't add the same device to the list again
I feel stupid because I dont understand how to download the dependencies or cloning repository.
Impeccable taste in vtubers!
yay thank u
Thank you !!!
Do you think there's more room of improvements for the Subtitler like making it use DeepL and enhancing the way it transcribe audio (it cuts way often unless you perfect the config for the timeouts)? It's a great way to understand Japanese and it could be used variously in different media such as animes, japanese podcasts, games, and more.
Yup there's definitely ways to improve it, I think the most glaring issue is that whisper is unable to handle simultaneous requests at once asynchronously, causing it to drop some requests.
And stream
hidden gem video
How do i run the command at 0:41?
i really want to stay at your house in the background ;-;
i love u thanks very much
i will do this for educational purposes
wait wha-
Uhhh Idk if you noticed but at 10:41 we can see your auth token in the command line underneath the black bar. Hopefully you've reset it already.
Muchisimas gracias
well it is work for me, thanks g.
can you send proof?
I really wanted to try this but I got an error every step so I dropped it but nice video
because it doesn't work lmao
Thanks bbe
Awesome
This man will bring a wave of waifus into the world. Sugoi!!!
Any plans to train your own ai? To possibly create a project similar to Neuro-sama, an ai vtuber created by a British dude to play osu, but recently started streaming chating with viewers and sometimes interact with her own creator.
hey man
love your videos..
i was wondering... did you try incorporating the NLTK module in python ?
to convert English to Japanese so that people who have some low end systems can still run the program when they have no access to the cloud or something
just my suggestion as i use the NLTK module a lot.
Oh this is the first time I have heard of this module haha, what is your experience of the accuracy of the translations and also the formality?
@@sociallyineptweeb
So im just a beginner in python😅.. i don't know as much as you lol😂. But in my experience of using NLTK(natural language toolkit) the accuracy of the sentence may not be as good as compared to the setup you've given us (it's really awesome btw!) but its fine for some low end pc's or people who just don't like the cloud for some reason. And since it is a python module in itself , i am assuming that it would run a little bit faster given that we aren't using DeepL(the middle man you mentioned in your first video about this waifu voice changer).
anyway im just a dude who knows the basics of python and some app building skills lol, i don't know too much but ig the NLTK wouldn't make a huge difference other that sacrificing accuracy and gaining a lil bit of performance .
looking forward to learn more from your videos btw and keep up the good content man!!
btw this is my first hearted comment... 😁😁
Good luck!
nice video ! Time to study english to speak japanese
So im a little confused to start. Are you running that initial command in Command Prompt, Powershell, Python, or docker?
In the final step, when i use the comand docker compose up -d i get this error: Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy'
nvidia-container-cli: initialization error: WSL environment detected but no adapters were found: unknown any idea? Q_Q
THIS VID IS AMAZING> HAVE A BELL
I just love the fact there are subtitles because we wont be able to hear him
What's the easiest way to test whether its working or not? Should I be hearing the waifu voice playback to me? or would it only play to the others in game? @11:54, what IDLE are you using in this part to see this code? Or where would I see that error if it popped up? I followed along pretty well but I feel like I'm missing something.
Edit: I've been trying to follow everything through windows command prompt. I think this is the reason I'm getting errors. Not sure where the best place to run this is.
mega based sheesh
Yey finally
4:25 When I change the output device to voicemeeter aux input, the subtitler works but I lose the sound from the application. I don't really know what I'm doing so if someone could help me out that would be greatly appreciated
I was in fact not being sarcastic at all, not even a little bit 😁
Cool vid bro, just wandering whats the minimum requirements to run all of these programs? (RAM etc)
Awesome work man, love the attention to detail :)
I wondered would there be a way to specify my second graphics card, to handle both of the programs while i use my main one for playing the game?
Did it work? Can you help me?
This is amazing! But I'm dumb so I'll try my best to follow😅
My brain went into fire [ It's fine : ) ]
I am determined to do this even though I am literally struggling really bad at the start I will do this no matter what no matter the time it takes
Update #1 it took me 26 minutes to do the first 2 minutes of the video not looking good but I will do this
update #2 I am learning a lot rn my brain is overheating I will go get some rest
@@L3gitXD. goodluck
@@ooostashooo4541 ty I will get this
@@L3gitXD. ask if you need help
also keep updating
dude making like 500k view video and telling my brother may find out that im making one
I hope you can sell this idea or make your own company out of this idea you've made. I want this to be integrated with all the games MMORPG I know... so badly (T v T)
Hey, thanks a lot for the setup guide. Other than some minor issues, which I managed to google or just reading through discord which I managed to fix. I managed to get it working through using google colab.
Got a question though, when using it on discord, do you have to switch the input to cable in order to use voicevox? As far as I'm aware, that seems to be the case, but it won't pick up my mic that way. And I would have to switch it back to mic input for others to hear my actual voice. Is this just a limitation that I can't get around?
Uh yah that's kinda a limitation haha
@@sociallyineptweeb haha I see. Your editing from when you tested this on discord made me think you might have some alternate methods. Maybe if you could route your mic input into the virtual cable perhaps? Though that would mean have its own problems
where can i open this 7:12? git terminal? or what apps and where and how?? :))
Been doing duolingo for over 200 days now, and I think it is helpful.
What is not helpful is being middle age with a brain who can't learn.
That Brain Fog is fr fr 😢
11:07 The whisper file returns a couple of URL. Which one do I paste in the env file?
I will wait for the UI.exe version 😅
time to fulfill my destiny as a female vtuber
did you even try the setup lol
Hi, I have a problem: When using Colab, I can't access the voicevox site where I can see the ID of the speakers because I either don't understand how or "replacing the url with the one from the colab" does not work :c
Aside from that, I love this project
This
Yo Ai-chan i've come to learn the way of creating AI waifu voiced translator. Then again i might not have the Power of gaming PC
is it normal for my whisper and voicevox tooking too long to start its just starting for 20 mins still not started
same
Same for me as well
lol same
Try giving them a priority though Task bar
Helped for me since I have way too many background operations