ardha27 / ai-song-cover-rvc

All in One Version : Youtube WAV Download, Separating Vocal, Splitting Audio, Training, and Inference Using Google Colab

AI-Song-Cover-RVC

All in One Repository: Youtube WAV Download, Separating Vocal, Splitting Audio, Training, and Inference Using Google Colab or Kaggle.

Make Sure To Leave A Star If This Repo Was Helpful :)

Tutorial (Indonesian)

RVC Colab Free Tutorial (Training & Inference)

• https://www.youtube.com/watch?v=oTum0CIGwns

AICoverGen Colab Free Tutorial (Inference)

• https://www.youtube.com/watch?v=Nd2b-MEynAg

Download YouTube WAV and Splitting Audio

• 

Training Without UI/Gradio, Prevent Banning (Colab Free)

Read this tutorial

Download Training Assets

• 

Run Training (Run Upper One First)

• 

AICoverGen Mod by Me & Hina (Colab Free)

• 

AICoverGen by SociallyIneptWeeb (Without UI, Colab Pro Only)

• 

Training V2 and Youtube Audio Download & Splitting Audio Combined by MinatoIsuki (With UI, Colab Pro Only)

• 

ai-song-cover-rvc's Issues

Web link v2

No link is provided when running RVC2

Copy result training to drive

malah keluar log ini mas, dan file yg di copy ke drivenya g lengkap

cp: cannot stat '/content/Retrieval-based-Voice-Conversion-WebUI/logs/sucro/added_.index': No such file or directory
cp: cannot stat '/content/Retrieval-based-Voice-Conversion-WebUI/logs/sucro/trained.index': No such file or directory
cp: cannot stat '/content/Retrieval-based-Voice-Conversion-WebUI/logs/sucro/total_*.npy': No such file or directory

The Modded AICoverGen (for Colab Free) Broke

The other cells worked fine but "Run WebUI" keeps getting an "ImportError" all of a sudden. What happened?

AttributeError while inferencing

It says something like "NoneType" has no attribute "dtype"

ModuleNotFoundError: No module named 'utils', Google Colab RVC Crepe v2 + persistent storage / autosave

Error in 4 from utils import backups

---

ModuleNotFoundError Traceback (most recent call last)
in <cell line: 4>()
2 import threading
3 import time
----> 4 from utils import backups
5
6 #@markdown #Click here to run the GUI

ModuleNotFoundError: No module named 'utils'

---

NOTE: If your import is failing due to a missing package, you can
manually install dependencies using either !pip or !apt.

To view examples of installing some common dependencies, click the
"Open Examples" button below.

RVC directory not created

Hello,

When executing the "Cloning Github" cell (section 2.2), I run into this error:
cp: cannot stat '/content/drive/MyDrive/RVC/weights/*': No such file or directory
The reason is the RVC directory has never been created.

What are those "weights" and what is supposed to generate them?

Thanks,
Kevin

Run web gradio

link gradionya ga muncul bang untuk yang versi v2

Gagal memuat audio sample / Failed to load audio sample

Halo bang, ane lagi coba training model di TrainingV2_NoUI.ipynb. Tapi malah gagal muat data audio sample buat training, padahal ane cek ulang semua file audio gak ada yang salah sama sekali. Dan juga file path nya juga udah bener dari google drive

Apakah ini ada hubungannya sama file KiwKiw nya atau ada masalah lain?

Split The Audio into Smaller Duration Before Training

mohon maaf yang bagian keduanya error karena typo gitu itu kenapa ya??

Loss Plot

Halo kak, aku mau lihat grafik loss untuk training rvc, kira kira aku bisa langsung nambahin load tensorboard ga ya di kodingan kakak? thankyouu!

Not Kaggle banning Gradio?

Sometimes Gradio did generate public link, but the code is automatically get Ctrl+C then disconnects from runtime.

Can't Train Feature Index

Good afternoon, I ran into this error: (as a newbie, i just don't know)

Trying doing kmeans 397811 shape to 10k centers.

NameError Traceback (most recent call last)
in train_index(exp_dir1, version19)
32 big_npy = (
---> 33 MiniBatchKMeans(
34 n_clusters=10000,

NameError: name 'MiniBatchKMeans' is not defined

During handling of the above exception, another exception occurred:

NameError Traceback (most recent call last)
1 frames
in train_index(exp_dir1, version19)
42 )
43 except:
---> 44 info = traceback.format_exc()
45 logger.info(info)
46 infos.append(info)

NameError: name 'traceback' is not defined

RVC Colab issue..

Gak tau masalahnya di colab atau RVC atau notebook nya.
Ada yg ngalamin google colab( free)selalu disconnect pas traning juga inference gak lebih dari 10min.
Jadi sudah gak bisa lagi pakai yang free , gak tau sih yg pro .belum nyoba.
Ada solusinya gak ya?
Atau adakah alternatif Colab yg free. Yg bisa rvc ? .thx

Model cannot download from Google Drive

Hello there,

it's been a long time since i did AI Covers.
I had to reload my Model voice from Google drive and it seemed to work, because it said "done".
But when i go to Inference after and refresh the site. "Model Voice" is empty.
Did i do anything wrong?

Looking forward for your help

The Weeknd - Hold Your Heart

Can you make it with The Weeknd AHTD Live vocals

Stuck at 'separating Vocals from Instrumentals'

Hi,

I've been using your AICoverGen Mod (Colab Free) and had no issue... until today, it's stopped working for some reason. I can run all cells and get through the WebUI, but whenever I want to generate AI cover, it's stuck at separating vocals from instrumentals. I checked the Colab and it says "Failed to create CUDAExecutionProvider".

Rekomendasi dataset

bang untuk dataset voice rekomendasinya berapa?
50-100 cukup kah atau ada rekomendasi lain ?

There's no pth file in the result file

I've watched the tutorial and follow step by step. But there's no pth file in my drive, I've tried it a few times and the result still the same. Only npy and index file

cp: cannot stat '/content/KiwKiw/assets/weights/ModelName.pth': No such file or directory
/content/drive/MyDrive/Model/ModelName
zip warning: name not matched: ModelName.pth
adding: added_IVF3572_Flat_nprobe_1_ModelName_v2.index (deflated 7%)
adding: total_fea.npy

Can't do AI Cover anymore

Hello there,

since a couple of days it doesn't work anymore. How can i solve this issue. Thanks in advance :)

the de-reverb does not work.

in All in One Inference by SociallyIneptWeeb
the de-reverb does not work, the voice still with reverb so the output not good.

Resume training ?

Hi,

How can I resume training on colab TrainingV2_NoUI.ipynb ? I know it is possible on some other RVC colabs but I'm not sure about this one.

Unzip File From Drive Failed

Hello, i have a problem on Training without Gradio on "Unzip File From Drive". I think the KiwKiw folder is missing

unzip: cannot find or open /content/drive/MyDrive/assets/KiwKiw.zip, /content/drive/MyDrive/assets/KiwKiw.zip.zip or /content/drive/MyDrive/assets/KiwKiw.zip.ZIP.
[Errno 2] No such file or directory: '/content/KiwKiw'
/content
ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt'

Error saat ingin convert Inferencing

bang pas saya convert error terus

Traceback (most recent call last):
  
File "/usr/local/lib/python3.10/dist-packages/gradio/routes.py", line 414, in run_predict
    output = await app.get_blocks().process_api(
  File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1323, in process_api
    data = self.postprocess_data(fn_index, result["prediction"], state)
  File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1257, in postprocess_data
    prediction_value = block.postprocess(prediction_value)
  File "/usr/local/lib/python3.10/dist-packages/gradio/components.py", line 2531, in postprocess
    processing_utils.audio_to_file(sample_rate, data, file.name)
  File "/usr/local/lib/python3.10/dist-packages/gradio/processing_utils.py", line 163, in audio_to_file
    data = convert_to_16_bit_wav(data)
  File "/usr/local/lib/python3.10/dist-packages/gradio/processing_utils.py", line 177, in convert_to_16_bit_wav
    if data.dtype in [np.float64, np.float32, np.float16]:
AttributeError: 'NoneType' object has no attribute 'dtype'
Traceback (most recent call last):
  File "/content/Retrieval-based-Voice-Conversion-WebUI/infer-web.py", line 146, in vc_single
    if_f0 = cpt.get("f0", 1)
NameError: name 'cpt' is not defined

Traceback (most recent call last):
  File "/usr/local/lib/python3.10/dist-packages/gradio/routes.py", line 414, in run_predict
    output = await app.get_blocks().process_api(
  File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1323, in process_api
    data = self.postprocess_data(fn_index, result["prediction"], state)
  File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1257, in postprocess_data
    prediction_value = block.postprocess(prediction_value)
  File "/usr/local/lib/python3.10/dist-packages/gradio/components.py", line 2531, in postprocess
    processing_utils.audio_to_file(sample_rate, data, file.name)
  File "/usr/local/lib/python3.10/dist-packages/gradio/processing_utils.py", line 163, in audio_to_file
    data = convert_to_16_bit_wav(data)
  File "/usr/local/lib/python3.10/dist-packages/gradio/processing_utils.py", line 177, in convert_to_16_bit_wav
    if data.dtype in [np.float64, np.float32, np.float16]:
AttributeError: 'NoneType' object has no attribute 'dtype'

Feature Extraction RMVE problem

Bang, ada masalah pas mau feature extraction sebelum training make RMVE

Loading rmvpe model
f0fail-9-/content/Retrieval-based-Voice-Conversion-WebUI/logs/Raisu/1_16k_wavs/115_1.wav-Traceback (most recent call last):
File "/content/Retrieval-based-Voice-Conversion-WebUI/infer/modules/train/extract/extract_f0_print.py", line 126, in go
featur_pit = self.compute_f0(inp_path, f0_method)
File "/content/Retrieval-based-Voice-Conversion-WebUI/infer/modules/train/extract/extract_f0_print.py", line 89, in compute_f0
self.model_rmvpe = RMVPE(
File "/content/Retrieval-based-Voice-Conversion-WebUI/infer/lib/rmvpe.py", line 590, in init
ckpt = torch.load(model_path, map_location="cpu")
File "/usr/local/lib/python3.10/dist-packages/torch/serialization.py", line 791, in load
with _open_file_like(f, 'rb') as opened_file:
File "/usr/local/lib/python3.10/dist-packages/torch/serialization.py", line 271, in _open_file_like
return _open_file(name_or_buffer, mode)
File "/usr/local/lib/python3.10/dist-packages/torch/serialization.py", line 252, in init
super().init(open(name, mode))
FileNotFoundError: [Errno 2] No such file or directory: 'assets/rmvpe/rmvpe.pt'

RVC_2_Training error generate and error splitting on WebUi

I got " --- Logging error ---
Traceback (most recent call last):
File "/usr/lib/python3.10/logging/init.py", line 1100, in emit
msg = self.format(record)
File "/usr/lib/python3.10/logging/init.py", line 943, in format
return fmt.format(record)
File "/usr/lib/python3.10/logging/init.py", line 678, in format
record.message = record.getMessage()
File "/usr/lib/python3.10/logging/init.py", line 368, in getMessage
msg = msg % self.args
TypeError: not all arguments converted during string formatting
Call stack:
File "/content/Retrieval-based-Voice-Conversion-WebUI/infer-web.py", line 30, in
from infer.modules.uvr5.modules import uvr
File "", line 1027, in _find_and_load
File "", line 1006, in _find_and_load_unlocked
File "", line 688, in _load_unlocked
File "", line 883, in exec_module
File "", line 241, in _call_with_frames_removed
File "/content/Retrieval-based-Voice-Conversion-WebUI/infer/modules/uvr5/modules.py", line 13, in
config = Config()
File "/content/Retrieval-based-Voice-Conversion-WebUI/configs/config.py", line 34, in wrapper
wrapper.instance = func(*args, **kwargs)
File "/content/Retrieval-based-Voice-Conversion-WebUI/configs/config.py", line 58, in init
self.x_pad, self.x_query, self.x_center, self.x_max = self.device_config()
File "/content/Retrieval-based-Voice-Conversion-WebUI/configs/config.py", line 121, in device_config
logger.info("Found GPU", self.gpu_name)
Message: 'Found GPU "
I Try to run audio split on webui cannot work and try to generate ai cover with other sample wav got error.

New Song AI Cover

Hey, does anyone know when it'll be possible again to create new AI Cover? It's so sad that it has been removed by google. Is there any other way? Because it's standing there that it has been removed for free Users. Can i become a premium user to use it again?
I love your work and i've got a lot of fans i don't want to disappoint.

Error when create ai cover

Traceback (most recent call last):
File "/content/AICoverGen/src/main.py", line 291, in song_cover_pipeline
voice_change(voice_model, main_vocals_dereverb_path, ai_vocals_path, pitch_change, f0_method, index_rate, filter_radius, rms_mix_rate, protect, crepe_hop_length, is_webui)
File "/content/AICoverGen/src/main.py", line 201, in voice_change
rvc_infer(rvc_index_path, index_rate, vocals_path, output_path, pitch_change, f0_method, cpt, version, net_g, filter_radius, tgt_sr, rms_mix_rate, protect, crepe_hop_length, vc, hubert_model)
File "/content/AICoverGen/src/rvc.py", line 150, in rvc_infer
audio_opt = vc.pipeline(hubert_model, net_g, 0, audio, input_path, times, pitch_change, f0_method, index_path, index_rate, if_f0, filter_radius, tgt_sr, 0, rms_mix_rate, version, protect, crepe_hop_length)
File "/content/AICoverGen/src/vc_infer_pipeline.py", line 571, in pipeline
self.vc(
File "/content/AICoverGen/src/vc_infer_pipeline.py", line 424, in vc
npy = np.sum(big_npy[ix] * np.expand_dims(weight, axis=2), axis=1)
IndexError: index -1 is out of bounds for axis 0 with size 0

If you want to do manual seperate ;

If you want to do manual seperate ;
It can be integrated into the sound obtained at the model output or the input sound. There is a seperate code inside but it can be used if you want to do it more manually.
I think it is more effective than the one in colab

from audio_separator import Separator

Initialize the Separator with the audio file and model name

seperate_target_audio = "/content/1.wav" #@param {type:"string"}
separator = Separator(seperate_target_audio, model_name='UVR_MDXNET_KARA_2')

Perform the separation

primary_stem_path, secondary_stem_path = separator.separate()

print(f'Primary stem saved at {primary_stem_path}')
print(f'Secondary stem saved at {secondary_stem_path}')

!Tensor error info

The tensor errors I got during Extract Features and model building are fixed as follows ;
!pip install numpy==1.21.6
!pip install tensorflow==2.12.0

Error when trying to run Web

I tried to run web, but it gave error in infer-web.py. Please fix it.

Files for 48k sample rate training aren't there... (Training/Trainer V2)

When downloading RVC V2, only the 40k sample rate files are downloaded.

#RVC V2
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/D40k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o D40k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/G40k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o G40k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/f0D40k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o f0D40k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/f0G40k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o f0G40k.pth

This makes the script output some error whenever I try to select the 48k sample rate.

As of now, when I re-checked the lj1995/VoiceConversionWebUI repository, there are more files now.

Not too sure why the 48k sample rate files are not downloaded, maybe because of the limitations of V2? Or maybe they just add it recently?
To fix the problem, we can just download everything from the pretrained_v2 directory.

#RVC V2
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/D32k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o D32k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/D40k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o D40k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/D48k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o D48k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/G32k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o G32k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/G40k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o G40k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/G48k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o G48k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/f0D32k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o f0D32k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/f0D40k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o f0D40k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/f0D48k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o f0D48k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/f0G32k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o f0G32k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/f0G40k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o f0G40k.pth
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/pretrained_v2/f0G48k.pth -d /content/Retrieval-based-Voice-Conversion-WebUI/pretrained_v2 -o f0G48k.pth

I am going to try this myself to see if it works. If it does, I will let you know!

EDIT: Sure enough, no errors! Time to train the Astolfo model with ✨ pure high-quality audio ✨ :D

training

bang kalo 300 epoch butuh berapa menit training nya?

eror rmvpe

yang rmvpe eror bang nggak bisa di gunain

Getting "Button is not defined" error

NameError Traceback (most recent call last)
in <cell line: 46>()
47 if 'end preprocess' in f.read():
48 clear_output()
---> 49 display(Button(description="\u2714 Success", button_style="success"))
50 else:
51 print("Error preprocessing data... Make sure your dataset folder is correct.")

NameError: name 'Button' is not defined

I have reduced my dataset size to an hour all split up into 5-10 second files. Previously when my data set was too big, I was told both minibatch and traceback were both not defined before, but now it's button.

The colab I'm using is the easyGUI📱- 3/09/24

Hubert / ContentVec

Halo kak, maaf izin tanya, untuk metode yang digunakan di RVC ini apakah menggunakan Hubert atau ContentVec ya kak?