Skip to content

Latest commit

 

History

History
73 lines (64 loc) · 3.03 KB

README.md

File metadata and controls

73 lines (64 loc) · 3.03 KB
Raw

PlaylistGPT

PlaylistGPT is a Python script designed to enhance your Spotify playlist experience by categorizing songs based on their lyrics. Built using the Syrics API for fetching playlist info and song lyrics and the GPT-4 framework provided by GPT4FREE for natural language processing.

Requirements

To use PlaylistGPT, you need:

  • Python 3.x
  • Access to Spotify Developer Console for the sp_dc key
  • Required Python packages (g4f, syrics)

Usage

  1. Setup:

    • Obtain your Spotify Developer Client (SP_DC) key and save it in SP_DC_KEY.txt.
    • Install required packages: pip install -r requirements.txt.
    • Configure the GPT-4 Free provider in the script if needed.

  2. Running the Script:

python3 playlist_gpt.py --verbose
  1. Menu Options:
  • 1. Get sp_dc token first: Retrieve SP_DC token from Spotify Developer Console.
  • 2. Get lyrics: Fetch lyrics for tracks listed in links.txt.
  • 3. Categorize lyrics: Analyze and categorize the downloaded lyrics.
  • 4. Export Playlists: Export categorized playlists to separate text files.
  • 5. Restart Session: Clear session and reset process.
  • 6. Clear cache: Clean the cache directory.
  • 7. Run all: Execute steps 2 and 3 consecutively.
  • 8. Complete reset: Reset system including cache and session data.

Example

Input

Consider the following example input with 10 Spotify track links in links.txt:

- https://open.spotify.com/track/4uY9DiDYULeTvV1FsYFrAW
- https://open.spotify.com/track/1BxYaRKQzdaikFV2xJNpxd
- https://open.spotify.com/track/0KSwFEkVh0KCBR0TwqBVen
- https://open.spotify.com/track/0TeJBV1JebreflQy2yb315
- https://open.spotify.com/track/3wS0JIapaGqM2zqsRi3p86
- https://open.spotify.com/track/5rAnfPGKZGyRBzI5kcAaqB
- https://open.spotify.com/track/5qzepeW1CAsX5JhgNudQuP
- https://open.spotify.com/track/6rdqVFwrsavgoEzV4G3nqU
- https://open.spotify.com/track/7rxzL7rKUVa7epgDFjecGW
- https://open.spotify.com/track/75bTohS26lM1nbgDutXCoc

Output

After categorization, the playlists are organized as follows: These can then be pasted into a new spotify playlist

Playlist "arriba":

- https://open.spotify.com/track/3wS0JIapaGqM2zqsRi3p86
- https://open.spotify.com/track/5rAnfPGKZGyRBzI5kcAaqB
- https://open.spotify.com/track/5qzepeW1CAsX5JhgNudQuP
- https://open.spotify.com/track/6rdqVFwrsavgoEzV4G3nqU

Playlist "tristes":

- https://open.spotify.com/track/0KSwFEkVh0KCBR0TwqBVen
- https://open.spotify.com/track/0TeJBV1JebreflQy2yb315
- https://open.spotify.com/track/7rxzL7rKUVa7epgDFjecGW
- https://open.spotify.com/track/75bTohS26lM1nbgDutXCoc

Playlist "belicos":

- https://open.spotify.com/track/4uY9DiDYULeTvV1FsYFrAW
- https://open.spotify.com/track/1BxYaRKQzdaikFV2xJNpxd

Disclaimer

This script uses external APIs and services, and its functionality may be subject to changes in these services. Use it responsibly and in compliance with the terms of service of the respective platforms. Use at your own risk. Do not use spotify data to train AI models. This is just NLP i think. :o