beatles

Beatles annotations - musical dataset exploration

About the dataset

http://isophonics.net/content/reference-annotations-beatles

Manually annotates songs of Beatles

• song structure - segments + labels
• keys - segments + labels
• chords - segments + labels
• chroma features

Time map visualization

https://districtdatalabs.silvrback.com/time-maps-visualizing-discrete-events-across-many-timescales

Multi-scale visualization of discrete events originating from physics of chaotic dynamic systems. Basically scatter of time-to-previous vs. time-to-next event.

$ ipython notebook

Open time_map_of_chord_events.ipynb in the Jupyter (http://localhost:8888/).

Chord labels to pitch class segments as features for further ML

classification_based_on_chords

• Chord labels parsed to binary pitch class sets via https://github.com/bzamecnik/chord-labels and added to the dataframe.
• It can be used to:
  • better analyses and visualizations
  • key classification based on chord contexts
  • chord classification based on chroma features
  • chord imputation and prediction
  • etc.

The classification in the notebook is really not complete...