If 13:00 Failed/Error use this kzread.info/dash/bejne/gJhmxI-GddzUcco.htmlsi=hNPdhytwgjXQHeLj Try another way for AI cover: cutt.ly/KeebmdbW

Ah, I remember watching your tutorial like 5 months ago, and I did my ai cover, but 'cause of the study and school, i abandoned doing that.. Now I returned, and when I couldn't find your channel i was so concerned! gladly what u just changed your channel's name :D Thank you for the tutorials!! Ill try them out now

as always thank you so so much! you're such an inspo since im just starting to study compsci

mantap bang gokil, nice info

Terimakasih bang.

makasih banyak ya bang tutorial nya. btw ada yang punya saran lain ga ya, supaya hasil cover ai dari modelnya nya bagus natural gitu suara nya. soalnya aku dah nyoba beberapa kali masih kayak ada suara robot kaku gitu, apa yang salah atau jelek di sample acapella nya atau model nya apa gimana dah

Thank you!

Thank you so much for this tutorial! I got it, how to do these AI covers, it's really impressed since last time I used it really... When I tried to make AI to song russian song (with english voice lines), i got the terrble and at the same time funny result. But now it's impressed! I tried again to sing bot the russian song AND OH MY GOD IT'S PERFECT, sounds almost like a real human! You do really good tutorials which helps to understood everything, thank you! :D

@ryusuikagaku

8 ай бұрын

Try to check my new video, it will give you more better result

@SunnySuzuki0su

8 ай бұрын

I have a question I came back to huggingface site after a day, logging at the same account, but when I click refresh to see my models, my models are not appearing, I checked, my models still in my drive google@@ryusuikagaku

@ifuckedmyselfwhenitalk

7 ай бұрын

go back to white space

In the dataset we can put one audio file or can we put all the audios in the folder ? Thanks for help

Bang kenapa web ui nya error code nya ya?? ada masalah di google colab sepertinya..mohon diperbaiki ya, tiap kali running code web ui nya, error import error; cannot import name documentation

bang plis jawab kalo audionya dr yt 14 menit an, cache gpu nya ga di centang, epoch 200 berapa lama bang prosesnya? ini belom aku proses, kalo kira² berjam² ga jadi😂

wah makasi bangg pengen coba jg

11:50 min kok pas di train model dan copy training result to drive malah error?

Kak yang di huggingspace kenapa pas di duplicate balik lagi2 ke akun yang running ya, gak building. Gak disuruh buat akun juga

This thing actually used to be a little faster, only took at most around 10 minutes to generate a 5 minute audio. Now a 5 minute like that audio took approx. 15 minutes. It occured after i restarted the space after a while of inactivity.

bang apa ada cara buat inference di komputer lokal? kebetulan PC udah mumpuni

Kalau mau hapus model yg dah diupload di hugging face gimana bang?

bang kenapa saya tidak dapat terhubung ke backend GPU ?

Muncul eror mulu kak pas bagian konvert cover lagu nya tah knp gtu mulu ya

Bagus bang

tq bg

Bang..pas mau di download model..dia gk bisa..dia tulis cannot download from private link padahal gw udah set everyone with link

Bang, training epochnya gak bisa dilanjutin kah? 200 epoch lanjut ke 500 epoch gak bisa kah?

Why does Aesthetic RVC Inference HF fail when I get to this step?

Bang kalau error train model dan copy training result to drive gimana solusinya ?

Bang ini kenapa pas mau convert ai voice di hugging face, ada tulisan error dipojokan atas

Lapor bang, inference yang di huggingface error semua. Apa ada niatan buat diperbaiki?

Kok di aku tulisannya audio was not properly selected or doesnt exist bang? Pdhl udh ikutin stepnya persis

i followed everything until 17:15 somehow theres **Error: None** that keeps showing up. what a bummer. idk how to fix it

padahal file Zip nya udah dibuka bang, tapi masih tetep statusnya privat link, Gimana ya bang?

udah ngikutin cuma pas di cek di google drive nya gak ada tuh folder dataset,,drive saya tetep aja kosong

bang mau nanya ko dari kemarin module not found error trus ya biasanya lancar aja

saran gua sih mending make Ultimate Vocal Remover sih buat misahin vocal dan instrument klo pke yg huggingface lama bet dan juga kurang gitu bagus buat misahin vocal dan instrument, klo pke uvr lebih bersih dan cepat juga

@ryusuikagaku

8 ай бұрын

Ni uvr built in kzread.info/dash/bejne/gJhmxI-GddzUcco.html

Bang punya ku kok terjadi kesalahan saat mengunduh yang di saat Download model itu

Aku udah cari-cari yang metode rvc 1 ternyata udah gaada padahal aku khattam itu hiks. Sayang sekali aku lupa caranya jadi belajar ini baru lagi 😓

Assalamu'alaikum bang buat tutorial bagaimana cara agar bisa buat lagu original sendiri pake ai mas please ya 👍👍👍👍👍

Q: Kenapa ganti nama channel bang? A: Gpp, pengen rebranding sama sesuatu yang fresh aja

@waddo9997

8 ай бұрын

sebelumnya nama channelnya apa bang?

@ryusuikagaku

8 ай бұрын

Ardha Pina Ch.@@waddo9997

@princessputridj1699

7 ай бұрын

Bang mau tanya, apa cara abang ini masih bisa di pakai? Soalnya setau ku bukannya code AI voice udah di banned google colab?

@ryusuikagaku

7 ай бұрын

@princessputridj1699 ini pake trik biar ga kedetect

@AzzahraNabila16

6 ай бұрын

Bang kok aku training, model zip nya malah gk ada

kak kalo mau ubah model yang udah saya train ke zip itu index nya atau apa nya yah yang harus diubah, terus ada opsi pemilihan zip sama 7z kak itu juga pilih yang mana kak?

@ryusuikagaku

8 ай бұрын

Select npy, pth, sama indexnya. Trus masukin kedalam satu zip. Pilih zip

Bang kenapa pas mau upload model di Google Drive selalu error "An error occurred downloading" begini mulu, padahal udah di publik in link nya

@ryusuikagaku

6 ай бұрын

kzread.info/dash/bejne/ip58squjgsTWgJc.html

How can I solve the problem that shows FileNotFoundError: [Errno 2] No such file or directory?😭

@ryusuikagaku

8 ай бұрын

Dont give space on audio/file/folder name

Bang, ini di gw doang atau emang huggingface error mulu dari kemarin ya bang?

Kalo lagu Indonesia di cover k bahasa Korea menggunakan ai bisa GK yaa

Oke bang gw coba

hello! Thanks for your very nice tutorial video! Could you help me please to change a singers voice with my voice (i have already seperate the singers music,singers vocals,and my voice) ? Because i am trying but i can not complete it :( Thank you !!!

@ryusuikagaku

5 ай бұрын

Where you find trouble?

@manolisgman5977

5 ай бұрын

I have got a message that it do not find some file. I think the name of file was: WiwI or something like that. I will try again today.I know that it is not easy for you but could you help me by some way? for example to send you some photos or video of the problem ? or could you help me to pass the problem by remote help? Thank you!@@ryusuikagaku

@manolisgman5977

5 ай бұрын

Hello! This is the problem message! python infer/modules/train/preprocess.py '/content/drive/MyDrive/dataset/me' 48000 2 '/content/KiwKiw/logs/me' False 3.0 python3: can't open file '/content/infer/modules/train/preprocess.py': [Errno 2] No such file or directory@@ryusuikagaku

Yang Hina gk bisa lagi ya bang???

Itu masih berfungsi RVC Aesthetic Interface

aku nyoba di HP kemarin gagal :V, ini mau nyoba lagi wkwkwkkw

Please help me! I'm almost at the end but there is a problem. When I place the link from youtube and i press download and separate, then it shows me the next message: Error - None. This happens to any link from KZread I put. Please tell me what i have to do! Thank you very much!

@ryusuikagaku

5 ай бұрын

Try this kzread.info/dash/bejne/gJhmxI-GddzUcco.htmlsi=hNPdhytwgjXQHeLj

@manolisgman5977

4 ай бұрын

@@ryusuikagaku Ok thank you!

bang spertinya saya sudah ada folder model di gdrive tapi kok ga bisa di download? di model interfacenya? bacaanya unknown error occured

@ryusuikagaku

7 ай бұрын

coba pake tutorial videoku yg setelah video ini

15:04 pisah suara

Kak cara nya ke wav gimana ya? Kok aku mp3 ?

Maaf kak aku mau tanya .tadi Aku sempat membuat Contoh Ai cover Song di hugging face aesthetic rvc tapi setelah di coba dan Di convert malah Error.cara Mengatasi nya gimana ya supaya Gak Error Lagi ?

@ryusuikagaku

7 ай бұрын

Pake ini aja lebih lancar kzread.info/dash/bejne/gJhmxI-GddzUcco.html

Bng klo udah punya model trs gw mau nambahin model lgi apakah perlu download traning assets???

@ryusuikagaku

7 ай бұрын

Ga perlu, cukup sekali

Bang gue kok gak bisa potong audionya ya bang? Kalau dataset karakter yang lain bisa kepotong. Itu gimana bang masih bisa di training kah? Soalnya gue lihat di folder dataset. File audionya cuman 1 dan utuh.

@ryusuikagaku

8 ай бұрын

File audiomu terlalu mepet/ngerap. Codinganku motong audio berdasarkan suara hening di audio. Hening = potong

bos q permisi mau tanya waktu msukin Separate KZread tracks lodingnya lama trus ujung2 nya eror gmna ya bos q buat ngtasinya

@ryusuikagaku

5 ай бұрын

durasi audionya brp lama

Kenapa dibagian download seperated audio di hugging face, lama banget proses nya Panjang durasi audio yg download itu sekitar 3 menit

@ryusuikagaku

8 ай бұрын

Karena disana prosesnya cuma pakai GPU, coba ini lebih cepet kzread.info/dash/bejne/gJhmxI-GddzUcco.html

Harus dari laptop ya kak.. klo dari hp bisa gak kak?

Maaf bang kenapa RAM nya tiba2 naik pas training epoc

The RVC inference link worked very well, however sometimes I just kept getting the disconnected error message even though I haven't leave the tab. Randomly I would also get the connection errored out message during convert. And other times I would just get this strange "Error: None" message.

@ryusuikagaku

8 ай бұрын

Did you check the logs for more detailed error?

@thanh2425

8 ай бұрын

@@ryusuikagaku I might try.

@thanh2425

8 ай бұрын

@@ryusuikagaku Also another thing I forgot to tell you is that everytime the disconnected and the connection errored out message pops up, it immediately stops converting & I had to do it all again. This is really annoying especially when I'm trying to convert a long audio

@ryusuikagaku

8 ай бұрын

@@thanh2425 how long is your audio? Try to split it if it too long, we are using free huggingface with limited resources

@thanh2425

8 ай бұрын

@@ryusuikagaku Less than a minute

Bang ada tutor gabungin model RVC atau kombinasi/gabungin 2 suara karakter gak bang? Biar jadi suara baru yang unik.

@ryusuikagaku

8 ай бұрын

Otw

@satoshidarikotamasara910

8 ай бұрын

Ok bang, kalau bisa sekalian bang bahas AI yang bisa ubah bahasa. Kali aja kan ada cara yang bisa ubah lagu ke bahasa lain pakai AI.@@ryusuikagaku

Halo teman saya, saya punya masalah di awal pekerjaan, saya menggunakan file audio dari Google Drive, tetapi tidak menerima pengerjaan (splitting) hanya menerima pekerjaan (terpisah), dan kemudian sebelum bekerja, tetapi masalahnya adalah tidak mengirim file di Google Drive, itu hanya membuat folder audio dan tidak membuat file audio, apa solusi untuk masalah ini.

@ryusuikagaku

5 ай бұрын

You can send it on English

Bang di huggingface, dibagian download model, udah ku salin url tapi statusnya "cannot get file fron this private link" Gimana dong

@ryusuikagaku

4 ай бұрын

huggingfacemu jangan di private

if i have trained a model up to 200 epoch and the result not good it is possibile to retrain the model based on these 200 epoch to 500 epoch to save time rather than repeat to train the model from scratch can you add this option in the code ?

@ryusuikagaku

7 ай бұрын

On current code, it's possible if still run the colab

@user-xr2fy4xi8y

7 ай бұрын

can you make a video for this option for exemple if i have trained the model up to 200 epoch but when im tryed i dont like it and i wan to retrain the model up to 500 epoch we dont have to repeat the process from scratch can you help ? @@ryusuikagaku

@ryusuikagaku

7 ай бұрын

@@user-xr2fy4xi8y if you still opening the colab, just change the epoch to 500 and rerun it, it will resume from 200

@user-xr2fy4xi8y

7 ай бұрын

Unfortunately, Colab occasionally disconnects on its own, forcing us to start the training process from scratch. This can be frustrating, especially if we don't have the option to upload the last trained model from our drive and resume the training from the last epoch. @ryusuikagaku @@ryusuikagaku

why it has error at first part ;-;

Bang adakah yg versi software gue cari gak nemu gak tau namanya Pkai aplikasi biar gk perlu website gitu?

@ryusuikagaku

7 ай бұрын

Blm pernah nemu

bang ini udah rmvpe? dan ini makan waktu berapa menit bang?

ada yang bisa bantu ga ni waktu mau salin ke bagian resources di hugginface pasti bakal kaya gini "Cannot get file from this private link"

@ryusuikagaku

6 ай бұрын

kzread.info/dash/bejne/ip58squjgsTWgJc.html

Kalau Ada update an Di Hina mod nya ,upload lagi ya bang🙏🙏

Saat mau download model di Aesthetic RVC, sudah ku share linknya seperti di tutorial, yaitu siapa saja yang memiliki link (viewer) lalu saat ku salin, dan ku tempel ke web nya. Tulisannya "Cannot get file from this private link"

@zeonplaych

5 ай бұрын

Foldernya juga udah diizinkan buat semuanya

@zeonplaych

5 ай бұрын

Ok bang dah ketemu masalahnya, ternyata belum login hehe. Lupa kalo pake akun lain

kalau udah punya RCV dari tutorial yang dulu bisa digunakan atau training ulang lagi ? soalnya RCV model lama gk ada zip dan isinya agak beda

@ryusuikagaku

8 ай бұрын

Di zip sendiri aja pake 7zip

@galehjati

8 ай бұрын

@@ryusuikagaku tapi jumlah file nya tidak sama gpp ?

@ryusuikagaku

8 ай бұрын

@@galehjati yg penting ada pth sama added.index

Kalau datasetnya paling idealnya berapa menit ya bang? 7 menit itu kalau di training 200-500x itu udah bagus? Aku ada dataset 1jam'an, dan itu bikin trainingnya lama, kalau 7-10 menit udah bagus atau ada rekomendasi idealnya berapa

@ryusuikagaku

8 ай бұрын

Dah bagus segitu

@rayhanjm6108

8 ай бұрын

@@ryusuikagaku siap, makasih bang

@yoshitakaputra7519

8 ай бұрын

@@rayhanjm6108 150 - 200 cukup sih, gua aja datasetnya 27 menit full suara nyanyi tanpa instrument dan suaranya bagus

bang huggingfacenya yang sekarang bisa dipakai juga ternyata sedangkan yang gradio versi collab malah harus error terus

@ryusuikagaku

4 ай бұрын

errornya apa, kirim ke discord

bang kalo compute units nya habis gimana bang harus beli yang pro kah

@ryusuikagaku

8 ай бұрын

Ngga, aku tanpa compute units kok

Ko prosesnya lama banget ya bang, hampir setengah jam nunggu gjd jd setelah itu koneksi error mulu. Gw sudah buat 3 voice model cuma pas bagian cover di rvc nya lama banget dan selalu koneksi error :( .

@ryusuikagaku

6 ай бұрын

kzread.info/dash/bejne/gJhmxI-GddzUcco.htmlsi=44ziK1e8Uutrkpkb

bang mau tanya kalau kita udh pernah nge-train model, apakah bisa memakai yang sudah sebelumnya dan kalau bisa. bagaimana cara menggunakannya?

@ryusuikagaku

8 ай бұрын

Tinggal di zip trus masukin drive atau huggingface

@chavezzx

8 ай бұрын

​@@ryusuikagakucara ubah jadi zip nya gimana kak

@ryusuikagaku

8 ай бұрын

@@chavezzx pakai software 7zip

@chavezzx

8 ай бұрын

​@@ryusuikagakubuatin tutorial nya dong kak ga ngerti nih

@ryusuikagaku

8 ай бұрын

@@chavezzx di google banyak, ga nyampe semenit ngezip file

kak kenapa ya proses building di saya lama banget 45 menit lebih belum selesai?

@aaghardan2832

6 ай бұрын

ancrit aku cuma 5 menit

Dikirain kepencet SB,ternyata cuman ganti nama sama PP😂

Bang cara Cari model2 Suara Dimana? Kayak nyari furina,Sama karakter lain, Pls answer this😢

@ryusuikagaku

8 ай бұрын

www.weights.gg

@kujojotaro989-j9g

8 ай бұрын

Cara downloadnya gimana bang?

bang jadi klo 200 epochs nunggu 200 menit y bg training nya?

@ryusuikagaku

8 ай бұрын

Tergantung panjang datasetnya juga

macam mana nak tahu cache gpu dia lebih dari 10 minit bang? ikut minit dari youtube yang kita copy untuk datasets ke?

@ryusuikagaku

7 ай бұрын

Iya

@AzizoXme

7 ай бұрын

@@ryusuikagaku nasip gw pakai teori gw sendiri ikut minit di youtube😂

bang, aku udah ikutin caranya tapi ko di gdrive gaada file index sama npynya ya?

@ryusuikagaku

6 ай бұрын

Di hp?

Bang.. Mau tanya nih, kenapa setiap convert inference selalu "Error None" di logs nya kyk gini " AttributeError : 'NoneType' object has no attribute 'tobytes' " , mohon bantuanya bang 🙏

@ryusuikagaku

8 ай бұрын

Coba kirim ss logsnya di discord

@imfive505

8 ай бұрын

@@ryusuikagaku ok bang, besok saya kirim

@imfive505

8 ай бұрын

@@ryusuikagaku link discord invalid or has been expired bang

ini make suara apa aja bisa kan ya di training nya? ane soalnya punya masalah di suara nya jadi gak jelas makany ane terpaksa hrs bikin video make suara ai (elevenlabs) tp dikomplen trs terlaloe ai jadi solusi gua cmn voice changer kek gini

@ryusuikagaku

7 ай бұрын

Bisa, suara knalpot juga bisa

@dinarsyauqie8397

6 ай бұрын

​@@ryusuikagakubjir knalpot aowkaowk

stuck di awal loading bg

Bang ada ga sih kode untuk memisahkan lead vocal dan backing vocal? Itu akan sangat membantu Thanks!

@chavezzx

8 ай бұрын

di web xminus udah ada fitur nya

@yoshitakaputra7519

8 ай бұрын

pake uvr

ku nemu channel org cover ai suaranya bagus+bisa nada panjang, dia pake apa?

@ryusuikagaku

8 ай бұрын

Tanya orangnya 🙏😌

Kalo ngedit pake hp gabisa kah?😭😭

Bang buat tutor cara buat TTS baru bang soalnya kata abang kan TTS free bahasa indo gak ada yang bagus.

@ryusuikagaku

8 ай бұрын

yang bikin sendiri juga aku blm ada nemu cara yang bagus

I did all the steps but when im going to convert hugging space says error

@kurikei1999x

5 ай бұрын

edit: I uploaded the files manually and it worked! thank you, you solved my problem.

@mori5583

4 ай бұрын

what do you mean by uploading manually? i have the same problem and i hope you can help me @@kurikei1999x

bang mau nanya ini aku pake colab ada keterangan gini "ImportError: cannot import name 'set_documentation_group' from 'gradio_client.documentation' (/usr/local/lib/python3.10/dist-packages/gradio_client/documentation.py)" gimana bang?

@ramadhan2335

5 ай бұрын

Masukin command di mana2 cell setelah error pip install gradio_client==0.8.1 Run web ulang dan udah bisa ada link

Gua nyoba pakai 50 epoch dengan suara henya sebanyak 3 jam bagus bang, tips juga goggle sekarang berbayar kalau lebih dari 6 jam wkwk

@ryusuikagaku

5 ай бұрын

tinggal ganti akun gmail atau tunggu 24 jam kalau akun kena limit

Bang, kok di Train Feature Index error mulu ya? Tulisannya cuman [ModuleNotFoundError: No module named 'faiss']

@ryusuikagaku

6 ай бұрын

Yg diatas udah di run semua? Btw mending lanjut diskusi di discord

@kaykvlen5715

6 ай бұрын

Tadi coba di rerun lagi, kali ini erornya mulai dari bagian "Unzip File from Drive", ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. lida 0.0.10 requires kaleido, which is not installed. Apa perlu install ulang QQnya lagi ya?

Kalo pakai Local runtime pakai VGA komputer pribadi bisa nggak bang?

@ryusuikagaku

8 ай бұрын

Itu VRAM Google Colab 16GB. Kalau VRAMmu dibawah itu berarti bakal lebih lambat prosesnya

@suryagameplay9257

8 ай бұрын

@@ryusuikagaku jadi begitu. Saya terkadang sedikit kelas karena runtime yang terputus tiba tiba karena credit saya 0. Jadi saya memcari cara agar bisa memakai hardware yang ada di PC/Laptop

bang kok punyaku pas di train gaada tulisan train epoch gitu ya, terus pas udah selesai training di drive modelnya gaada pth nya

@ryusuikagaku

7 ай бұрын

Berarti ada yg error

@fikoo6090

7 ай бұрын

@@ryusuikagaku kira² itu error di apanya ya bang??

@ryusuikagaku

7 ай бұрын

@@fikoo6090 kirim log errornya ke discord

Bang pas gua copy link yang si google drive ko ga bisa ya pas pencet Download, ada bacaat cannot apa gitu

@dandipermadi1230

6 ай бұрын

udah bisa tapi ngomong ngomong berapa lama si preses separate youtube treak nya?

@ryusuikagaku

6 ай бұрын

Pake ini coba kzread.info/dash/bejne/gJhmxI-GddzUcco.htmlsi=E80J9vmy7VLuaKr0

Bagi link thumbnailnya bang

bg, itu training assets klo dijalanin filenya 2 kali, filenya jadi ganda ga??

@ryusuikagaku

8 ай бұрын

Dijalanin 2x utk apa?

@user-od6dw6pb1g

8 ай бұрын

@@ryusuikagaku jadi sekali aja bg?? Misal klo kita train Hu Tao jalanin itu, ntar klo pen train Nahida gak ush jalanin lgi??

@ryusuikagaku

8 ай бұрын

@@user-od6dw6pb1g ohh, iya sekali aja. Kecuali kalau mau update rvc terbaru (semisal ada)

@user-od6dw6pb1g

8 ай бұрын

@@ryusuikagaku ok bg makasih, fast respond five stars ⭐⭐⭐⭐⭐

audio was not properly selected or doesn;t exist ... bang

@ryusuikagaku

5 ай бұрын

Pas step menit brp

@ILHAM-tz2jv

5 ай бұрын

@@ryusuikagaku Last step bg convert, tapi keknya kendalanya di duplicate nya, jadi nge skip bagian yang ngebuild

@ILHAM-tz2jv

5 ай бұрын

dh bisa bang thx @@ryusuikagaku

bang, aku masih di tahap prepare dataset tapi kalo hasilnya di gdrive audionya gak ke split? masih sama dengan durasi audio original

@ryusuikagaku

8 ай бұрын

Rekaman suaramu terlalu mepet/ngerap. Codeku motong audio berdasarkan suara hening di audio

@hoshinoharu01

8 ай бұрын

@@ryusuikagaku min, aku coba nyari voicelines nya fischl. tapi ada suaranya Oz nya, audio yg kyk gitu gak bisa yah?

@ryusuikagaku

8 ай бұрын

@@hoshinoharu01 harus dibersihin suara Oznya

@hoshinoharu01

8 ай бұрын

@@ryusuikagaku oke makasih banyak min, berhasil tutor nya 🙏😁

@ryusuikagaku

8 ай бұрын

@@hoshinoharu01 tonton ini juga biar makin mantap kzread.info/dash/bejne/gJhmxI-GddzUcco.html

eror pas generate 17:30