Tracksy

👀 Visualize and understand how you listen to music.

• The same old song? Late at night? Only in summer?

• Tracksy aims to help you answer these questions. It helps you see your data without us seeing your data.

🦆 Powered by DuckDB WASM and Astro x React.

🚧 Currently only works with Spotify streaming data.

Project Structure

Inside of this project, you'll see the following folders and files:

/
├── app/
└── synthetic-datasets/

app/ is the front end. It's an Astro project using DuckDB for data storage. This means that your uploaded data stays with you, client-side. We don't want to help you understand your data without knowing your data. For developers or curious, take a look at app/README.md for more details on development.

synthetic-datasets/ is use to generate datasets to test the Tracksy application. For developers or curious, take a look at synthetic-datasets/README.md for more details on development.

Usage

⬇️ Download your data

You can use pre-generated datasets available on Hugging Face tracksy/synthetic-datasets. For example, download one of the files from the interface or with the following command:

# /!\ The size of this file is greater than 44Mb, other files are available.
curl -LO https://huggingface.co/datasets/tracksy/synthetic-datasets/resolve/main/datasets/spotify/streamings_500000.zip?download=true

Or use your own Spotify data:

1. Request your Spotify data on your Spotify account
   • Select "Extended streaming history"
   • Click on "Request data"
   • Confirm your request by clicking on Spotify's confirmation e-mail
2. 30 days or less later
3. Open the mail from Spotify and download files

🚀 Upload your data

Go to Tracksy and upload your file.

You can also deploy Tracksy on your own, just take a look at the documentation.

👀 Visualize your data

Once you've uploaded your file, you'll be able to see graphs about your data. 🎉

"Roadmap"

The project is currently under development. We plan to add visualizations as well as a simplified view to make understanding streaming data as accessible as possible. We also intend to support other data sources (Deezer, Funkwhale, etc.). See our issues.

Contributing

Contributions are welcome! Feel free to open an issue or a pull request (don't forget to consult our CONTRIBUTING.md).