@@theAIsearch Just 1 question, in the chapter [continue training model] do you tweak anything?, epochs? do you have to hit [train feature index], or just go to train model thanks
@@theAIsearch thanks for the video really helpful but sadly i got a problem after i downloaded the 4.7 GB INSTALL_Mangio-RVC-v23.7.0_INFER_TRAIN.bat the window closed and no 7-zip rar was appearing and i tried restarting my pc and yet i couldn't find the problem so i ended up re-downloading it again and again no 7-zip rar appeared i even re installed the latest rar in your link and tried again downloading the mangio-rcv but now it finishes the download so quickly as it is already done and yet no 7-zip to be found i also checked my dowloads every file in my laptop and i cant find it. ( my laptop isnt a mac its an asus tuf gaming with a GTX 1660 TI and an AMD ryzen 7)
@@theAIsearch Error. Did you have any idea ? Mangio-RVC-v23.7.0\infer-web py", line 300, in vc_single if_f0 = cpt.get("f0", 1) NameError: name 'cpt' is not defined
As somebody who doesnt know much about Ai - This is the perfect video 30 minutes ago I had no clue about AI voiceovers and how to do them. Now I can even train my own voice model ! Thank you for making it so easy to understand :)
@@theAIsearch While training model 17:15 Process Data not generating anything in output window . I have read the githib Installing the Dependencies. 1.Installed python3.9.8 is stable with this fork 2Installed dependencies in requirments.txt floder .pip install -r requirements.txt I tried everything I can for past two days but cant solve the problem
Thanks dude!! You're SOOO much more different that other creators who just recommend tools that aren't free but claim to be in their channel. You just gained a sub 🎉
I’ll most definitely try this one out. I’m pretty tired of using those watered down, limited and trial versions. Then when you decide to pay good money for something there’s no software, that’s accessed strictly through a browser, then realize for the really good features I have to pay more for a pro version.
Other than leaving a space in a file directory name this was smooth as could be, it's almost as if the AI are making your tutorials for you, great video!
Tips : if your files are in the rvc audios folder, you don't need to copy/paste/rename the target file. Just use the option below "Auto detect audio path and select from the dropdown:" and select your sound file in the list.
Aw man, I should’ve watched this video first!!! I paid $10 to use a website to separate vocals from the instrumental - I was so impressed, I used to do it by hand and it wasn’t nearly as good!!! But for FREE?? Hell yeah! I first got to your page through your video on training AI voice models through the RVC colab - my first test voice model is about halfway done so now I’m onto this video to take a peek at what I’ll be doing next!!! I’m a music fan/hobbiest, trading unreleased tracks with other fans, being into every little b-side and promo track ever done from my favourite artist… I just want to have some fun having my favourite artist “”cover”” some of my other favourite songs, and to be able to share them with my other music editing hobbyist friends. This is going to be awesome!! Thank you so much for your helpful videos!!!!! I was always the “computer geek”, but MAN I fell off as AI started up!!!! All the shit we used to play around trying to do for daaaays just for it to turn out shitty, you can automatically do in about an hour now!!!!!
This is literally THE BEST CHANNEL ABOUT AI, seriously, why did I only find it now?! Man, you are doing god’s work, considering the AI tool landscape is changing so quickly and all of the other RUclipsrs either make videos about outdated software, or straight up promoting shitty web services from some no-name companies.
Hello @ai-tools-search . Can u please tell what should be the epochs for a 10 minute clip, a 20 minute clip and a 30 minute clip for best optimized output (training) Please reply asap
They say you shouldn't have space in the folder's name. I had the same problem. I deleted the program, installed it again in right to the D:\RVC folder (so no complicated path), and turned out the Windows defence, then gave every access to the firewall... then it worked.
It didn't work for me because I didn't name the folder "RVC." I named it Vocal Cloning. Once I named it RVC, everything worked. Not sure if you experienced that or not
You are a god! I really have to praise you for this video! You really make a good tutorial, no bullshit dragging out the time, you are just talking about the subject in such compelling way. Thank you so much. I tried to check out some other videos about this, and they never got to the point, to much self promotion, they are talking about rvc and then suddenly a one minute promo for their new sweater they have bought cheap from a basement in india😅 You actually had a very clever promotion for subscribing since it was a part of your sound clip. I appreciate that. I will follow and will press like for all your future videos. Please continue with these videos! Have a great day🎉
Who ever created this web interface, is a complete brain wreck. Why not make buttons small and have basic BROWSE buttons to pick paths and files ? Why not make buttons smaller and in correct order ? Darn linux mind set.
Ohh, I'm the dummy that left too much silence in my clips, so I thought they didn't work. You'd think I'd never done anything like this before. Thanks for making.
Thank you for this guide! This is the best A.i. voice guide I found on youtube from 0 to 100. I was really surprised how well 2 - 3 minutes voice can be train and use effectively from your advise
Hey, I got a question... When I'm doing the voice conversion, I do the same thing as you in the video one by one, but when I hit the ,,Convert" button, an error sign appears and I can't do anything about that. What do you think could be causing the trouble?
Make sure that you use the correct extension name for the file that you're converting. To see it in windows go to the file explorer. Click View at the top > show > file name extensions.
17:16 I'm so curious why I can't process my data into the output information even follow your steps but you successed. Can you tell what's the problems that I face now?
I had the same problem. I deleted the program, installed it again in right to the D:\RVC folder (so no complicated path), and turned out the Windows defence, then gave every access to the firewall... then it worked.
hey, thanks for the tutorial!! just a small pothole, however: for some reason none of the models show up in the vocals/accompaniment separation section, would you happen to know why this is? im using macOS if that helps :)
3:24 is not internal or external command, executable program, or batch file Press any key to continue After this the command window closes and nothing happens
3:18 when I did this the file was just called *go-web* and said that 'runtime\python.exe' was not recognized as an internal or external command, operable program, or batch file. What do i doooooo?
Those stuck at phase 2 "Processing data" that print nothing as output: make sure the file path as no folders with space or special characters(only letters). For example: C:\Users\User name\Desktop\TrainModelSoftware\gura\ It will print nothing because the folder "User name" have a space in the name.
For people stucked in step 2, Its very important to set "number of CPU" according to your hardware, if you have 8 cores but you leave it as 11, it will stuck. in my case, i put 6 of my 8 cores, that fixed my problem.
An easy to to get the file/folder path is to hold SHIFT and RIGHT CLICK on the file/folder, and the option "Copy to Path" should show up. (if I'm not mistaken).
I'm commenting while the TRAIN file is downloading, so I may edit it after posting, but I am really quite thrilled to find a solution that does not require subscribing to some online service. I really prefer to do everything on my own, from my computer, without paying a monthly fee. It's the reason I don't use Photoshop anymore, for example. I see in your description that you don't recommend the GTX 1080. Sadly, that is exactly the GPU I have. I've been meaning to upgrade to an RTX 2080ti (pretty cheap on eBay in 2024) - which is what you are using, based on the 17:23 of the video, even though the description states you're using an RTX 4080 - but money's been a bit tight. I'll report back on whether this works or not, using my old GPU. I don't mind if it's slow. I can run it while I sleep or am at work. But if it doesn't work at all, or the results are negatively affected, I'll have to pick up some OT at work to get the RTX. Good luck on turning this side hustle into a full-time thing!! I totally subscribed just for the ability to not have to pay a subscription fee for this tech.
Yo bro i am facing some error like G:\Mangio-RVC-v23.7.0>runtime\python.exe infer-web.py --pycmd runtime\python.exe --port 7897 Traceback (most recent call last): File "G:\Mangio-RVC-v23.7.0\infer-web.py", line 79, in conn = sqlite3.connect('TEMP/db:cachedb?mode=memory&cache=shared', check_same_thread=False) sqlite3.OperationalError: unable to open database file G:\Mangio-RVC-v23.7.0>pause Press any key to continue . . .
This has come a long way since last i messed with it, so much easier to set up and there's a ton of models already made, i'm not sure i even need to train anything anymore, thanks for making me aware.
If anyone experiences an issue with "Process Data" not generating anything in output window, make sure you name the folder "RVC" in the very beginning. I didn't & nothing worked until I renamed it RVC lol
What do you mean? I did everyithing as the video says, with a folder named "RVC" and get no output after processing, just blank :(. Folder direction looks like this: D:\RVC\Mangio-RVC-v23.7.0\Jose-Lavát DB
Ok i don't understand what happened, but from the default direction "D:\RVC\RVC\datasets/" i just deleted datasets and renamed with my audio fold and worked, wtf
Thank you so much. you are amazing, unlike the other youtubers, when you say free, you mean it, just for that you are genuinely a star. thank you for sharing your knowledge, you are unique. bless you
You have to be patient. Sometimes it takes from one to two minutes after hitting the .bat file before it moves to the next section. So don’t worry, it’s coming.
would be awesome if it also had a feature to turn text into those voices. So you can just write what the AI should say, this way you skip the record and "upload" process
I haven't tested it out yet, but I'm just so impressed with how informative your video is! You weren't just like "just follow the directions and you'll be fine blah blah blah" you gave specific details and directions on how to locate certain files and where to put them and that is just so helpful! Praying this works on my old, slow, pain in the butt laptop of mine.
Unfortunately, nothing you showed in the video works on my new laptop. If I press the CONVERT button, the process continues spinning forever. It doesn't stop when the duration of the audio file is reached.
i have a problem, i did everything correctly but when i click process data nothing happens :( it shows me ValueError: invalid literal for int() with base 10
Very well put together tutorial. Found this last night and made tons of covers like crazy. It's actually insane how well some of these models perform. Now I'm gunna secretly record my friend in discord calls and create a model and make him say a bunch of dumb shit lmaooo
what do i do if this message pops up in output information : 训练结束, 您可查看控制台训练日志或实验文件夹下的train.log ? i start the training process, then this comes up mere seconds after, not having trained any epochs at all? how do i fix this?
Thank you so much , I tried the Collab one the other day , but it didn't work , it just shuts off , I wanted to install it locally , but didn't know how , and then your video got recommended to me since I'm subscribed to you , the most impressive thing is the fact that you trained a model speaking English and made it sing a Japanese song
@@theAIsearch in training tab , the " process data " is not working , it kept going for more than 6300 seconds so i stopped it , and i think it's not using my GPU at all even though it recognized it , if there is a solution please tell me , thanks
Thank you so much for this tutorial, it explains it very clearly and is super easy to understand. I trained it and added one of my audio recordings to the training I did and omg I had to laugh so hard at it as it was nothing like what I was expecting, but I will keep training it till it gets the voice somewhat right, this is a lot of fun that's for sure. Again thank you so much for this great tutorial it's perfect for people like me who have no coding skills and want to mess around with AI stuff. you earned my sub that's for sure mate :D
Question. At Enter the path of the training folder, I would get an error called ( invalid literal for int() with base 10: ) How do I fix that? I tried copying address and It still doesn't work
i dont have a mac but here's the steps from github: install MacOS_Only_M1_Chip_Recommended_Mangio-RVC-Fork-23.7.0-HOTFIX.zip. For inference, all that needs to be done is running the run.sh from the extracted zip folder, where it will install Python 3.8, Homebrew, and other dependencies for you automatically. M1 Macs are natively supported for GPU acceleration, and training should work if you choose to download the pretrained models.
I have a problem. I createcd my model's name, added the path, clicked process data and nothing happened at all. NOTHING. I just see "processing" for a brief second and then a huge blank space. If I go on with the tutorial, at the end it shows me "error" and i can't train the model. What can I do?
@@theAIsearch no, nothing. The error shows up only in the ending when i press train, no more info. Trough the whole process til training, when processing data, creating index etc. it does a loading for like a few milliseconds and then blank empty space, when i should see end process instead. I tried to close and open again, I disintalled and reinstalled everything, deleted files from my pc, but nothing happened. And i followed your tutorial step by step til the installation process (i then skipped forward trough the train section since is the one i need). Did i miss something?
@@canguroidrovolante I had the same problem and I solved it. Maybe the essence of your error is different, but try all the folders that lead to the path of your sample, do not put spaces, as it may fail and the path will be interrupted. I made all the folders without spaces and it worked!
@@Antonin1423 i also tried an other RVC and the message on training was that Cuda Gpu memory is full or something like that... I don't know if those two messages have something in common...
I’m wondering, will the audio/vocal separation quality be better than how UVR does it? I never really used the separation option in v1 RVC, but will try out the v2 version if it does.
Great video! Got an issue, when i press process data after it does not give me anything it just stops. In the command prompt it shows 'False'. Any tips?
Just let it run like 10-15 mins if you have low end PC like mine and maybe it will show a enable like python administrator thing then yes and it should do it
For the ultimate vocal remover. It has problems when there is an instrument that shares the same melody as the lyrics. When I used it on Banat Iskandariyah, an Arabic song, an instrument joins the vocals in the separated audio
Hi there. Somehow it doesn't work by me. The part with "Model Inference" and the "Train" part. When i convert by "Model Inference" down below it stands Error. I did excatly what you explained and did it a few times i even tried 2 different pth files. By the section "Train" when i push "process data" in "Output infromation" nothing happens. And as i said i did excatly what you explained and did it a few times. I don't get it. If you can help i appreciate it and ty. By the part "Train" when i look up at cmd it says: FileNotFoundError: [Errno 2] No such file or directory: 'E:\\RVC\\Mangio-RVC-v23.7.0/logs/Kenjiro/preprocess.log' But it is the right folder. cmd: runtime\python.exe trainset_preprocess_pipeline_print.py E:\RVC\Mangio-RVC-v23.7.0\Kenjiro 40000 8 E:\RVC\Mangio-RVC-v23.7.0/logs/Kenjiro Tsuda False Traceback (most recent call last): File "E:\RVC\Mangio-RVC-v23.7.0\trainset_preprocess_pipeline_print.py", line 25, in f = open("%s/preprocess.log" % exp_dir, "a+") FileNotFoundError: [Errno 2] No such file or directory: 'E:\\RVC\\Mangio-RVC-v23.7.0/logs/Kenjiro/preprocess.log'
Danke!
No reply lemme fix that
;-; huh
Not often you find a perfect tutorial - hack - descriptive - helpful video combo, but here we have one TYSM.
Thank you!
@@theAIsearch Just 1 question, in the chapter [continue training model] do you tweak anything?, epochs? do you have to hit [train feature index], or just go to train model thanks
Is this tutorial also for ai song covers
What is the solution for converting my own voice taking a long time?
@@abhaygholap3613 if u got a low spec computer then it will struggle with large files.
Cheers! Only non clickbait video on the topic! U deserve the world, my dude!
thanks!
@@theAIsearch thanks for the video really helpful but sadly i got a problem after i downloaded the 4.7 GB INSTALL_Mangio-RVC-v23.7.0_INFER_TRAIN.bat the window closed and no 7-zip rar was appearing and i tried restarting my pc and yet i couldn't find the problem so i ended up re-downloading it again and again no 7-zip rar appeared i even re installed the latest rar in your link and tried again downloading the mangio-rcv but now it finishes the download so quickly as it is already done and yet no 7-zip to be found i also checked my dowloads every file in my laptop and i cant find it. ( my laptop isnt a mac its an asus tuf gaming with a GTX 1660 TI and an AMD ryzen 7)
@@theAIsearch s Heur.BZC.ONG.Leopard.1.1BE12493 My antivirus detected it, Any answers?
@@theAIsearch Error. Did you have any idea ?
Mangio-RVC-v23.7.0\infer-web py", line 300, in vc_single
if_f0 = cpt.get("f0", 1)
NameError: name 'cpt' is not defined
@@opafritzsche Yea me too
As somebody who doesnt know much about Ai - This is the perfect video 30 minutes ago I had no clue about AI voiceovers and how to do them. Now I can even train my own voice model ! Thank you for making it so easy to understand :)
Glad I could help!
@@theAIsearch
While training model
17:15
Process Data not generating anything in output window . I have read the githib Installing the Dependencies.
1.Installed python3.9.8 is stable with this fork
2Installed dependencies in requirments.txt floder .pip install -r requirements.txt
I tried everything I can for past two days but cant solve the problem
Subscribed buddy....thanks...
Seen a genuine RUclipsr after a long time
Thanks for the sub!
Very clearly explained, excellent tutorial :D. Got my model training now and I can't wait to see how it turns out!
Thanks dude!! You're SOOO much more different that other creators who just recommend tools that aren't free but claim to be in their channel. You just gained a sub 🎉
Appreciate that! Thanks for the sub!
frr
exactly. THis dude gets it
I’ll most definitely try this one out. I’m pretty tired of using those watered down, limited and trial versions. Then when you decide to pay good money for something there’s no software, that’s accessed strictly through a browser, then realize for the really good features I have to pay more for a pro version.
Other than leaving a space in a file directory name this was smooth as could be, it's almost as if the AI are making your tutorials for you, great video!
hope you blew up man! This is very helpful!
Thanks!
Tips : if your files are in the rvc audios folder, you don't need to copy/paste/rename the target file. Just use the option below "Auto detect audio path and select from the dropdown:" and select your sound file in the list.
Thank you so much! This actually made the difference between it working and not working for me for some reason
life saver
lifesaver man xD
where do you find the auto detect audio path button
Aw man, I should’ve watched this video first!!! I paid $10 to use a website to separate vocals from the instrumental - I was so impressed, I used to do it by hand and it wasn’t nearly as good!!! But for FREE?? Hell yeah!
I first got to your page through your video on training AI voice models through the RVC colab - my first test voice model is about halfway done so now I’m onto this video to take a peek at what I’ll be doing next!!!
I’m a music fan/hobbiest, trading unreleased tracks with other fans, being into every little b-side and promo track ever done from my favourite artist… I just want to have some fun having my favourite artist “”cover”” some of my other favourite songs, and to be able to share them with my other music editing hobbyist friends. This is going to be awesome!! Thank you so much for your helpful videos!!!!! I was always the “computer geek”, but MAN I fell off as AI started up!!!! All the shit we used to play around trying to do for daaaays just for it to turn out shitty, you can automatically do in about an hour now!!!!!
glad it was helpful!
This is literally THE BEST CHANNEL ABOUT AI, seriously, why did I only find it now?! Man, you are doing god’s work, considering the AI tool landscape is changing so quickly and all of the other RUclipsrs either make videos about outdated software, or straight up promoting shitty web services from some no-name companies.
Thank you!
Hello @ai-tools-search . Can u please tell what should be the epochs for a 10 minute clip, a 20 minute clip and a 30 minute clip for best optimized output (training)
Please reply asap
Wow - Brilliant video with a super simple thumbnail - thank you so much
Bro nobody could help me until i find you! Subbed🎉🎉🎉🎉
thanks!
@@theAIsearchno problem
it is better then 90% of video in youtube that say " change your voice " - thanks man
PROCESS DATA at 17:13 doesnt work, nothing popping up in logs, what comes up in command prompt is ( value error invalid literal
They say you shouldn't have space in the folder's name.
I had the same problem. I deleted the program, installed it again in right to the D:\RVC folder (so no complicated path), and turned out the Windows defence, then gave every access to the firewall... then it worked.
yes, finally, it works! thanks! @@zoltanlunk1654
@@zoltanlunk1654 omg I just change name to the one that has no spaces in name (single word) and it worked Thank you!
It didn't work for me because I didn't name the folder "RVC." I named it Vocal Cloning. Once I named it RVC, everything worked. Not sure if you experienced that or not
You are a god!
I really have to praise you for this video!
You really make a good tutorial, no bullshit dragging out the time, you are just talking about the subject in such compelling way.
Thank you so much.
I tried to check out some other videos about this, and they never got to the point, to much self promotion, they are talking about rvc and then suddenly a one minute promo for their new sweater they have bought cheap from a basement in india😅
You actually had a very clever promotion for subscribing since it was a part of your sound clip. I appreciate that.
I will follow and will press like for all your future videos. Please continue with these videos!
Have a great day🎉
Thank you so much!
What is requirements to install and run it? I mean PC requirements.
amazing amazing tutorial i spent over a week and finally this is the only one that gives me some hope, thanks a lot!
Who ever created this web interface, is a complete brain wreck.
Why not make buttons small and have basic BROWSE buttons to pick paths and files ?
Why not make buttons smaller and in correct order ?
Darn linux mind set.
Ohh, I'm the dummy that left too much silence in my clips, so I thought they didn't work. You'd think I'd never done anything like this before. Thanks for making.
No worries!
Huge thank you for making this tutorial! It's very well put together and a huge help for so many of us interested in using this technology!
You're very welcome!
how time the feature extraction will take for gtx 1650?
@@theAIsearch
Thanks for this buddy. Earned a Sub. 🎉
Thank you for this guide!
This is the best A.i. voice guide I found on youtube from 0 to 100.
I was really surprised how well 2 - 3 minutes voice can be train and use effectively from your advise
Glad it helped!
the only tutorial that's coherent and works
What they don't tell you in this program is the first voice is free but you only have one use out of it
My guy wakes up and decides to help the community
Such a great tutorial :) Simple, detailed and the best part.. in the end it actually works! Thank you x3
Hey, I got a question...
When I'm doing the voice conversion, I do the same thing as you in the video one by one, but when I hit the ,,Convert" button, an error sign appears and I can't do anything about that.
What do you think could be causing the trouble?
same problem, did u solve it?
@@davidebarbato4596 No, I tried everything, but nothing works
I switched from 24 hours of MP3 to a 3 minute wav file and everything started working.
I switched from 24 hours of MP3 to a 3 minute wav file and everything started working. @@davidebarbato4596
Make sure that you use the correct extension name for the file that you're converting. To see it in windows go to the file explorer. Click View at the top > show > file name extensions.
17:16 I'm so curious why I can't process my data into the output information even follow your steps but you successed. Can you tell what's the problems that I face now?
Same problem 😢
I answered how to solve it, but for some reason mt comment was removed (WTH?).
I had the same problem. I deleted the program, installed it again in right to the D:\RVC folder (so no complicated path), and turned out the Windows defence, then gave every access to the firewall... then it worked.
dude thank you so much! this is what I was looking for
No problem!
hey, thanks for the tutorial!! just a small pothole, however: for some reason none of the models show up in the vocals/accompaniment separation section, would you happen to know why this is? im using macOS if that helps :)
3:24 is not internal or external
command, executable program, or batch file Press any key to continue
After this the command window closes and nothing happens
3:18 when I did this the file was just called *go-web* and said that 'runtime\python.exe' was not recognized as an internal or external command, operable program, or batch file.
What do i doooooo?
did you find a solution?
this is so perfect thank you!! and its my first time doing this i dont even know about coding or anything but this is so helpful
You're welcome!
17:09 I got stuck when after i pressed Process Data the output box shows nothing:(
Same
make sure you dont use words like "voice" in either your audio extract or save filename. For some reason it doesn't like that
@@TheVision2 It still does not work
В названии эксперимента тоже пробелы убрать надо. Видимо, программа вовсе не любит пробелы
Make sure you have the \ at the end of the location
Thank So much , Love from Thailand Bro ❤❤
The Finest and Greatest Tutorial on RUclips❤
thanks!
Thanks, this was the only tutorial that worked for me.
You are welcome!
Those stuck at phase 2 "Processing data" that print nothing as output: make sure the file path as no folders with space or special characters(only letters).
For example:
C:\Users\User name\Desktop\TrainModelSoftware\gura\ It will print nothing because the folder "User name" have a space in the name.
hey man, just want to give you a big thumbs up. This tutorial is too good to be free.
Thank you!
For people stucked in step 2, Its very important to set "number of CPU" according to your hardware, if you have 8 cores but you leave it as 11, it will stuck. in my case, i put 6 of my 8 cores, that fixed my problem.
Can you help me? I have got Macbook M1 chip.. Having done downloading it for M1 chip, which file should I run?
An easy to to get the file/folder path is to hold SHIFT and RIGHT CLICK on the file/folder, and the option "Copy to Path" should show up. (if I'm not mistaken).
Thanks! just right click works for me (win 11)
Thanks! Great tip!
I'm commenting while the TRAIN file is downloading, so I may edit it after posting, but I am really quite thrilled to find a solution that does not require subscribing to some online service. I really prefer to do everything on my own, from my computer, without paying a monthly fee. It's the reason I don't use Photoshop anymore, for example.
I see in your description that you don't recommend the GTX 1080. Sadly, that is exactly the GPU I have. I've been meaning to upgrade to an RTX 2080ti (pretty cheap on eBay in 2024) - which is what you are using, based on the 17:23 of the video, even though the description states you're using an RTX 4080 - but money's been a bit tight.
I'll report back on whether this works or not, using my old GPU. I don't mind if it's slow. I can run it while I sleep or am at work. But if it doesn't work at all, or the results are negatively affected, I'll have to pick up some OT at work to get the RTX.
Good luck on turning this side hustle into a full-time thing!! I totally subscribed just for the ability to not have to pay a subscription fee for this tech.
Thanks! 2080ti is enough for this. good luck!
@@theAIsearch FYI, your UVR5 video has been removed. If you know what terms it violated, and can re-upload it, that'd be awesome! Thanks!
hi when i click the go-web bat file it just says this "location\Mangio-RVC-v23.7.0_INFER_TRAIN>pause
Press any key to continue . . ."
same here
he just replies to the comments that praise him... wow.
3:23 i open the .bat , press the key and nothing happens
You have to wait an hour , took me a long time
same it just says press any key
run as admin
I guess you need python to run that
Yo bro i am facing some error like G:\Mangio-RVC-v23.7.0>runtime\python.exe infer-web.py --pycmd runtime\python.exe --port 7897
Traceback (most recent call last):
File "G:\Mangio-RVC-v23.7.0\infer-web.py", line 79, in
conn = sqlite3.connect('TEMP/db:cachedb?mode=memory&cache=shared', check_same_thread=False)
sqlite3.OperationalError: unable to open database file
G:\Mangio-RVC-v23.7.0>pause
Press any key to continue . . .
I have a similar issue also about python.exe
me too, same error did you get it to work after all?
@@marcwuestenhagenDownload the infer version
@@marcwuestenhagenthis how I have fix it
Only infer
bro idk how can I tell u thank you but tbh u are my guardian!❤❤❤❤
This has come a long way since last i messed with it, so much easier to set up and there's a ton of models already made, i'm not sure i even need to train anything anymore, thanks for making me aware.
indeed, it has gotten so much easier & better in only a few months
If anyone experiences an issue with "Process Data" not generating anything in output window, make sure you name the folder "RVC" in the very beginning. I didn't & nothing worked until I renamed it RVC lol
What do you mean? I did everyithing as the video says, with a folder named "RVC" and get no output after processing, just blank :(.
Folder direction looks like this:
D:\RVC\Mangio-RVC-v23.7.0\Jose-Lavát DB
Ok i don't understand what happened, but from the default direction "D:\RVC\RVC\datasets/" i just deleted datasets and renamed with my audio fold and worked, wtf
For me, the 7z file isnt getting saved. and i cant follow the steps after that cus it just wont appear. can u help
I get stuck at the first line when opening go-web. what can I do?
I have the same question
I waited for 5 minutes i think, and it opened by itself
@firechiII then? what did you do after that? i have same problem :(
but the thing i noticed is it's just go-web not with the .bat like on the video
I have an issue when I go to open the web page by double clicking the go-web it just brings me to the transcript and doesn't further the process
did you ever find a solution? same problem here
@@illjp5823 unfortunately no I didn’t
Same problem here
@@jaykumarprajapati4627 hey bruh i got the solution
it took like 2-3 mins for me to open so wait a bit
You just got a new subscriber, thanks so much! Legendary W
Thanks for the sub!
Brilliant tutorial and funny interpretation :) Thank you!
Wow you do the best tutorial for AI, thank you so much
You're very welcome!
its worked for you ?
i cant train with all correct steps
thanks a lot, you are definitely at another level, thanks men.
Thank you so much. you are amazing, unlike the other youtubers, when you say free, you mean it, just for that you are genuinely a star. thank you for sharing your knowledge, you are unique. bless you
You're welcome!
nvm, i just restarted this app, and now it is working. thanks :)
Great!
When I start the go-web file, it just stays one line of code and it doesn't start anything, any tip how to fix it?
You have to be patient. Sometimes it takes from one to two minutes after hitting the .bat file before it moves to the next section. So don’t worry, it’s coming.
@@1000vox thank you for the advice. At the end of all started
@@sashkosashko2586 Glad to hear it!
would be awesome if it also had a feature to turn text into those voices. So you can just write what the AI should say, this way you skip the record and "upload" process
it has that. see this: ruclips.net/video/jh83ViDUvJs/видео.html
I haven't tested it out yet, but I'm just so impressed with how informative your video is! You weren't just like "just follow the directions and you'll be fine blah blah blah" you gave specific details and directions on how to locate certain files and where to put them and that is just so helpful! Praying this works on my old, slow, pain in the butt laptop of mine.
Thanks & good luck!
It works?
@@theAIsearchI can’t even get past the first step. Opening the go-web.bat file.
how long did your extraction take ?
a lot
Unfortunately, nothing you showed in the video works on my new laptop. If I press the CONVERT button, the process continues spinning forever. It doesn't stop when the duration of the audio file is reached.
Sir, I don't have nvidia graphics card, how can I use train model? Please help
just input 1 in the GPU section
How do you pause an AI voice model training in progress, or can you even pause it? Do I just close it up around the last saved epoch?
close all windows. but note that you will lose your progress
@@theAIsearch All the progress or from the last saved epoch?
i have a problem, i did everything correctly but when i click process data nothing happens :(
it shows me ValueError: invalid literal for int() with base 10
are there any spaces in the folder path/name? it was showing up for me too until i changed the names to have no spaces or special characters
@dodalinfr as in, any character other than the basic English alphabet and numbers ex: ășöêç
Very well put together tutorial. Found this last night and made tons of covers like crazy. It's actually insane how well some of these models perform. Now I'm gunna secretly record my friend in discord calls and create a model and make him say a bunch of dumb shit lmaooo
Thanks. For covers, check out this: ruclips.net/video/o6ywe3EDLDM/видео.html The last one is very convenient
what if it didnt open a newwindow in your browser (does this work on mac)
AI doesn't work on Mac
For some reason the convert step never finishes even for a 5 seconds track. Nonethess, the video is very informative
mine too :(
It could be because your GPU isn't powerfull enough
@@Hend8yeah I have integrated graphics and I am facing the same issue.😭🤣
So how do you solve it? I tried a 10 seconds audio file and it was at 250 sec loading and counting. Should i let it for an hour or somethin?
@@Beunnu Bro I left my 30 second audio file for an hour but it didn't worked it's just kept going.
what do i do if this message pops up in output information : 训练结束, 您可查看控制台训练日志或实验文件夹下的train.log ?
i start the training process, then this comes up mere seconds after, not having trained any epochs at all? how do i fix this?
I could not get this to work, it kept running into run errors
Great video, bro! Thank you.
You're welcome!
Thank you so much , I tried the Collab one the other day , but it didn't work , it just shuts off , I wanted to install it locally , but didn't know how , and then your video got recommended to me since I'm subscribed to you , the most impressive thing is the fact that you trained a model speaking English and made it sing a Japanese song
Thanks for the sub! Yeah, English to Japanese works because English covers most/all the tones of Japanese
@@theAIsearch in training tab , the " process data " is not working , it kept going for more than 6300 seconds so i stopped it , and i think it's not using my GPU at all even though it recognized it , if there is a solution please tell me , thanks
@@ilyesilfest7641 what error message do u see?
@@theAIsearch sorry , there is no error message , it just keeps going like i said , I don't know where the problem is , thank you for your time
sorry , i got the folder name with space in it , i fixed that , but in feature extraction it says : no-feature-todo
thanks
Extraordinary 🤩
Thank you so much for this tutorial, it explains it very clearly and is super easy to understand.
I trained it and added one of my audio recordings to the training I did and omg I had to laugh so hard at it as it was nothing like what I was expecting, but I will keep training it till it gets the voice somewhat right, this is a lot of fun that's for sure.
Again thank you so much for this great tutorial it's perfect for people like me who have no coding skills and want to mess around with AI stuff.
you earned my sub that's for sure mate :D
Thanks for the sub! Glad you're having fun with it!
Question. At Enter the path of the training folder, I would get an error called ( invalid literal for int() with base 10: ) How do I fix that?
I tried copying address and It still doesn't work
Also have this problem
This is the best tutorial I've ever watched, loved how its well put together. Thank You so much, I am subscribing
Thanks for subbing!
@@theAIsearch WELCOME, GREAT CONTENT. KEEP GOING
Great tutorial! Thanks a lot 😊
That's not the reason Colab stops your runtime. It's because RVC uses a web UI to operate, and Colab's rules prohibit the use of remote UIs.
Interesting, thanks for sharing!
nah they banned my notebook even without web UI
best vid ever bout this subject!
can you please make a detailed guide on how to install it in mac
i dont have a mac but here's the steps from github:
install MacOS_Only_M1_Chip_Recommended_Mangio-RVC-Fork-23.7.0-HOTFIX.zip. For inference, all that needs to be done is running the run.sh from the extracted zip folder, where it will install Python 3.8, Homebrew, and other dependencies for you automatically. M1 Macs are natively supported for GPU acceleration, and training should work if you choose to download the pretrained models.
I have a problem. I createcd my model's name, added the path, clicked process data and nothing happened at all. NOTHING. I just see "processing" for a brief second and then a huge blank space. If I go on with the tutorial, at the end it shows me "error" and i can't train the model. What can I do?
Do you see any more info about the error
@@theAIsearch no, nothing. The error shows up only in the ending when i press train, no more info. Trough the whole process til training, when processing data, creating index etc. it does a loading for like a few milliseconds and then blank empty space, when i should see end process instead. I tried to close and open again, I disintalled and reinstalled everything, deleted files from my pc, but nothing happened. And i followed your tutorial step by step til the installation process (i then skipped forward trough the train section since is the one i need). Did i miss something?
@@canguroidrovolante I had the same problem and I solved it. Maybe the essence of your error is different, but try all the folders that lead to the path of your sample, do not put spaces, as it may fail and the path will be interrupted. I made all the folders without spaces and it worked!
Train Index says: ERROR No such file or directory. And it's true! There is no .index file in logs file. Does anyone knows why?
Same
@@Antonin1423 i also tried an other RVC and the message on training was that Cuda Gpu memory is full or something like that... I don't know if those two messages have something in common...
try this ruclips.net/video/tnfqIQ11Qek/видео.html
@@theAIsearch thanx i am gonna try it!
@@theAIsearch mannn it works!
A huge process but finally were able to do it successfully 😭😭😭😭😩😩😩🥰
This was very helpful, thanks for this tutorial. It was really helpful
Does it worked...?
@@disko8291 yep but time consuming
I’m wondering, will the audio/vocal separation quality be better than how UVR does it? I never really used the separation option in v1 RVC, but will try out the v2 version if it does.
This uses UVR so it's similar. However for UVR, you can use a lot of different algorithms to get more custom results
Amazing!! Thx from Brazil!!!
Welcome!
Great video!
Got an issue, when i press process data after it does not give me anything it just stops.
In the command prompt it shows 'False'. Any tips?
same problem here
Thank you so much. Great Video.
web-go does not open browser
Same
Just let it run like 10-15 mins if you have low end PC like mine and maybe it will show a enable like python administrator thing then yes and it should do it
Thanks You So Much. Good Teacher! :D
You are welcome!
Hi Bro, can we install face swap also on local computer (bcz google colab not running often) ? if so please let us know in the next video.
Good job brah!!!
Subscribbled!!!
Thanks!
@@theAIsearch No! Thank YOU!!!
I love you, thanks a bunch!
You're welcome!
how to use it in Discord?
do u mean in realtime?
@@theAIsearchyea does it work realtime?
Yep - is there a way to feed it the voice in realtime and get like a virtual audio cable which can be plugged in into discord as an input
If youre having problems with "Process data" just use the original "audios" folder. Delete the two audio files and use what you need
For the ultimate vocal remover. It has problems when there is an instrument that shares the same melody as the lyrics.
When I used it on Banat Iskandariyah, an Arabic song, an instrument joins the vocals in the separated audio
it sounds like your using MDX-UVR which does this. i suggest VR Architecture
wow you deserve 1 mil sub!
wow thank you!
Hi there. Somehow it doesn't work by me. The part with "Model Inference" and the "Train" part. When i convert by "Model Inference" down below it stands Error. I did excatly what you explained and did it a few times i even tried 2 different pth files.
By the section "Train" when i push "process data" in "Output infromation" nothing happens.
And as i said i did excatly what you explained and did it a few times. I don't get it. If you can help i appreciate it and ty.
By the part "Train" when i look up at cmd it says:
FileNotFoundError: [Errno 2] No such file or directory: 'E:\\RVC\\Mangio-RVC-v23.7.0/logs/Kenjiro/preprocess.log'
But it is the right folder.
cmd:
runtime\python.exe trainset_preprocess_pipeline_print.py E:\RVC\Mangio-RVC-v23.7.0\Kenjiro 40000 8 E:\RVC\Mangio-RVC-v23.7.0/logs/Kenjiro Tsuda False
Traceback (most recent call last):
File "E:\RVC\Mangio-RVC-v23.7.0\trainset_preprocess_pipeline_print.py", line 25, in
f = open("%s/preprocess.log" % exp_dir, "a+")
FileNotFoundError: [Errno 2] No such file or directory: 'E:\\RVC\\Mangio-RVC-v23.7.0/logs/Kenjiro/preprocess.log'
Same as with me the zip file a long list of errors files
I love how you say you can unzip a 7z file with Winrar, but don't mention 7zip, the free open-source software that was the progenitor of the filetype.