Create AI Cover Songs Online for Free

howto create ai covers
Spread the love

AI generated cover songs are trending on the internet. Cover songs with Trump’s voice or Modi’s voice , sounds cool right? If you are wasting your time searching for the software which generates AI cover songs , then this article is for you . You can change the voice of any song to your preferred artist’s voice .

***Points to keep in Mind before Starting the creation:

  1. Not available on Mobile devices.( in mobile, you can try covers.ai website, but they have limited voices and it gets too much time for your creation)
  2. Need High speed internet.
  3. Don’t use long audios.
  4. Most Important: You have to be quick , very quick. As, you don’t have the premium plan, You’ll get a limited time to edit your audio. So, be quick.

Create AI Cover Songs: Step by Step:

Part 1: Choose a song: 

You can select any song to create its cover. But, it is recommended that, if your preferred artist(Whose voice you want to get as the output) is a Male, then choose song which was sung by a male singer. And choose a female voiced song, if you want to get the result in a female voice. 

Part 2 : Voice Isolation:

You have to split the voice and instrumental of the song. Simply use this website called: Acapella Extractor. https://www.acapella-extractor.com/

You can create acapella of two files per day using this website. Upload the song and wait for some time it will generate your vocal file of the song, you will need the instrumental file also . You will get another option to download the isolated instrumental file below. You will need that later . After downloading 2 files , you can now close the tab.

Part 3: Trim the Audio ( If needed):

Dont use audios more than 2 minutes(recommended, but u can try).

Part 4: Choose the output voice:

If you want to create AI generated cover songs of Donald Trump, Arijit Singh, Donald Duck, Modi or any other person, There are already thousands of premade voice models.

You have to select the voice model URL of your preferred person from this spreadsheet.

https://colab.research.google.com/corgiredirector?site=https%3A%2F%2Fdocs.google.com%2Fspreadsheets%2Fd%2F1tAUaQrEHYgRsm1Lvrnj14HFHDwJWl0Bd9x0QePewNco%2F

Also, join this Telegram group for new voice models: https://t.me/cubemusic99

( You’ll get a welcome message , Skip it by tapping “ Automatically sorted by Downloads(Latest )“)

 Go to the spreadsheet and search for your preferred voicemodel by clicking Ctr. + F [for example: we are going with ariana grande ]

1. Tap Ctrl.+ F

2. Type in the search bar you preferred voice ( for me , it’s Ariana Grande)

3. It will highlight the tab , if that person’s voice is available to clone. If not, then select any other voice 🙁

4. Once you get the result, you’ll get  single or multiple URL/URLs beside the tab. We need just one. There might be urls from gdrive/huggingface/mega. All works the same. So, just copy the URL.

Step4: Head to google collab:

Now, you have to be real quick for these next steps. You should close other tabs. 

Click this link to go to the Google collab page: https://colab.research.google.com/drive/1Gj6UTf2gicndUW_tVheVhTXIIYpFTYc7?usp=sharing#scrollTo=Y13Eh9r_g8f-

You’ll get a page like this 👇

You’ll find An option like this: “Step 1. Install (it will take 30-45 seconds)

And there is a button like play button beside the text ”cells hidden”. Tap this play button and it will take 30 to 45 seconds to create the environment. A pop up will appear, click “Run Anyway”. 

Once Step 1 is completed , you have to paste the URL of your voice model (which you copied from the spreadsheet/ telegram) in the step 2 ( Download the model ) , and hit the play button of step 2. Once, step 2 is completed head to the step 3 and hit the play button of step 3 (“Start the GUI, then open the public URL. It’s gonna look like this:”)

After some moments it will load some codes. And at the end of the code , you’ll get a Public URL ( The url beside the text ” Running on public URL”)

Click on the Public URL without thinking anything , cause, we are running out of time.

A new tab will be opened like this 👇

Now drag and drop the file which you just isolated from the acapella extractor website. It will show uploading .

*Do not hit the Convert button until the audio file is shown in the 2nd point ,i.e.” 2. Choose your audio( It will show Someguy named file in default ) sometimes it gets refreshed automatically and show your file.

But if you don’t see your file in the 2nd point, keep hitting the refresh button of the 2nd point, until your file appears at the 2nd point.

Now, just hit “ Convert “ button. It will show “Processing….” and your AI cover song will be ready within some seconds. Now simply download the file .

After some time your runtime will be over, but you got your Cover Song. Now, you can follow this process again and again and create your cover songs.

There is just one problem , that you are not getting enough time to edit your songs. Sometimes it will be failed , maybe you can not execute the process before the time.

But no worries, try again. If you have any doubt , do comment below.

Leave a Reply

Your email address will not be published. Required fields are marked *

This website uses cookies to ensure that you get the best experience.
Learn More
We only send notifications for Top Posts Ok No