Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/wightwick/fynesse

Pure-Python web app for music exploration with Spotify 🌱
https://github.com/wightwick/fynesse

Last synced: 14 days ago
JSON representation

Pure-Python web app for music exploration with Spotify 🌱

Awesome Lists containing this project

README 

        
# :control_knobs: fynesse



:seedling: Advanced recommendation and search web app for Spotify, written in pure Python.



:package: UI written in [Reflex](https://github.com/reflex-dev/reflex/), Spotify interaction via [Spotipy](https://github.com/spotipy-dev/spotipy)



![Screenshot](screenshot.png?raw=true 'fynesse UI')



## Features



- **Generate recommendations** based on seed tracks and artists from your library or search results + 5 tuneable parameters provided by rhe recommendation API

- **Browse** your Spotify library (liked songs, playlists, recently played) and see artist genres associated with each track

- **Fine-grained search** for artists and tracks — by name, genre, year

- **Play and queue new music** on your currently active Spotify device, or save recommendations to a playlist to lisen later



## Connecting to Spotify API and running the app



The core functionality of fynesse requires a connection to **Spotify's Web API**.



To set this up:

1. Go to the [Spotify developer dashboard](https://developer.spotify.com/dashboard/applications)

1. Click `Create app`

1. Enter a name and app description (these don't matter)

1. Enter `http://localhost:3000` as Redirect URI (note - needs to be the same port the reflex app is configured to run on)

1. Tick `Web API` under `Which API/SDKs are you planning to use?`

1. Agree to terms and click `Save`

1. Click `Settings` in the top right

    - You now can see your `Client ID` and `Client Secret`

1. Add your `Client ID` and `Client Secret` to [sp_secrets.py](sp_secrets.py)

1. You are now ready to run the app. Do so by running `reflex init` followed by `reflex run` in the [parent directory](/) (you will need the packages listed in [requirements.txt](requirements.txt) installed)

1. On first run, you will be prompted to log in to your spotify account. Do so!

1. The app should run, accessible at `http://localhost:3000` :sunglasses:



## Tips for use

- have a flick through your library for some bangers

- plant them and/or their artists into the seeds window

- germinate the seeds to produce a batch of tune recommendations

- experiment with the other generation parameters to fine tune your recommendations

- play/queue some/all of the recommended tunes

    - music is played via your active spotify device if you have one

    - **to play music from fynesse, you need to have a spotify client open somewhere - desktop, web, or mobile, and Spotify must consider it the 'active device'. To make a device active you can quickly play/pause from it**

- save recommended tunes to a playlist in your library to listen another time

- explore genres by showing them on your library and clicking to search

    - narrow down search to find similar tunes manually

- using a set of disparate seeds will produce a mixed playlist

- tunes can be played, queued or seeded from search results via each track's dropdown button