This is a set of scripts for copying "liked" songs and playlists from Spotify to YTMusic. It provides a GUI (implemented by Yoween, formerly called spotify_to_ytmusic_gui).
-
Install Python and Git (you may already have them installed).
-
Uninstall the pip package from the original repository (if you previously installed
linsomniac/spotify_to_ytmusic
):On Windows:
python -m pip uninstall spotify2ytmusic
On Linux or Mac:
python3 -m pip uninstall spotify2ytmusic
Start by creating and activating a Python virtual environment to isolate dependencies.
git clone https://github.com/AmidelEst/spotify_to_ytmusic.git
cd spotify_to_ytmusic
On Windows:
python -m venv .venv
.venv\Scripts\activate
pip install ytmusicapi tk
On Linux or Mac:
python3 -m venv .venv
source .venv/bin/activate
pip install ytmusicapi tk
To use the YouTube Music API, you need to generate valid credentials. Follow these steps:
- Log in to YouTube Music: Open YouTube Music in Firefox and ensure you are logged in.
- Open the Inspection Tool: Press
F12
or right-click and select Inspect to open the browser's inspection tool. - Access the Network Tab: Navigate to the Network tab and filter by
/browse
. - Select a Request: Click one of the requests under the filtered results and locate the Request Headers section.
- Toggle RAW View: Click the RAW toggle button to view the headers in raw format.
- Copy Headers: Right-click, choose Select All, and copy the content.
- Paste into
raw_headers.txt
: Open theraw_headers.txt
file located in the main directory of this project and paste the copied content into it.
Run the Script:
Execute the following command to generate the credentials file:
On Windows:
python spotify2ytmusic/ytmusic_credentials.py
On Linux or Mac:
python3 spotify2ytmusic/ytmusic_credentials.py
Important: After running this script, the authentication file will be created. When you launch the GUI in the next step, it will automatically detect this file and log in to YouTube Music without requiring manual input. You’ll see a log message confirming this:
File detected, auto login
The GUI will ignore the 'Login to YT Music' tab and jump straight to the 'Spotify Backup' tab.
Now you can use the graphical user interface (GUI) to migrate your playlists and liked songs to YouTube Music. Start the GUI with the following command:
On Windows:
python -m spotify2ytmusic gui
On Linux or Mac:
python3 -m spotify2ytmusic gui
Once the GUI is running, you can:
- Backup Your Spotify Playlists: Save your playlists and liked songs into the file
playlists.json
. - Load Liked Songs: Migrate your Spotify liked songs to YouTube Music.
- List Playlists: View your playlists and their details.
- Copy All Playlists: Migrate all Spotify playlists to YouTube Music.
- Copy a Specific Playlist: Select and migrate a specific Spotify playlist to YouTube Music.
It will go through your Spotify liked songs, and like them on YTMusic. It will display the song from Spotify and then the song that it found on YTMusic that it is liking. I've spot-checked my songs and it seems to be doing a good job of matching YTMusic songs with Spotify. So far I haven't seen a single failure across a couple hundred songs, but more esoteric titles it may have issues with.
This will list the playlists you have on both Spotify and YTMusic, so you can individually copy them.
You can either copy all playlists, or do a more surgical copy of individual playlists. Copying all playlists will use the name of the Spotify playlist as the destination playlist name on YTMusic.
To copy all the playlists click the copy
button, and wait until it finished and switched to the next tab
NOTE: This does not copy the Liked playlist (see above to do that).
In the list output, find the "playlist id" (the first column) of the Spotify playlist and of the YTMusic playlist.
The copy playlist will take the name of the YTMusic playlist and will create the playlist if it does not exist, if you start the YTMusic playlist with a "+":
Re-running "copy_playlist" or "load_liked" in the event that it fails should be safe, it will not duplicate entries on the playlist.
The function first searches for albums by the given artist name on YTMusic.
It then iterates over the first three album results and tries to find a track with the exact same name as the given track name. If it finds a match, it returns the track information.
If the function can't find the track in the albums, it then searches for songs by the given track name and artist name.
Depending on the yt_search_algo parameter, it performs one of the following actions:
If yt_search_algo is 0, it simply returns the first song result.
If yt_search_algo is 1, it iterates over the song results and returns the first song that matches the track name, artist name, and album name exactly. If it can't find a match, it raises a ValueError.
If yt_search_algo is 2, it performs a fuzzy match. It removes everything in brackets in the song title and checks for a match with the track name, artist name, and album name. If it can't find a match, it then searches for videos with the track name and artist name. If it still can't find a match, it raises a ValueError.
If the function can't find the track using any of the above methods, it raises a ValueError.
- Does this run on mobile?
No, this runs on Linux/Windows/MacOS.
-
How does the lookup algorithm work?
Given the Spotify track information, it does a lookup for the album by the same artist on YTMusic, then looks at the first 3 hits looking for a track with exactly the same name. In the event that it can't find that exact track, it then does a search of songs for the track name by the same artist and simply returns the first hit.
The idea is that finding the album and artist and then looking for the exact track match will be more likely to be accurate than searching for the song and artist and relying on the YTMusic algorithm to figure things out, especially for short tracks that might be have many contradictory hits like "Survival by Yes".
-
My copy is failing with repeated "ERROR: (Retrying) Server returned HTTP 400: Bad Request".
Try running with "--track-sleep=3" argument to do a 3 second sleep between tracks. This will take much longer, but may succeed where faster rates have failed.
Creative Commons Zero v1.0 Universal
spotify-backup.py licensed under MIT License. See https://github.com/caseychu/spotify-backup for more information.