Spotify Audio Features Dataset
Social Media and Posts
Tags and Keywords
Trusted By




"No reviews yet"
Free
About
This dataset presents the 100 most streamed songs of all time on Spotify, complete with their audio features extracted using the Spotify API. It serves to provide a clear understanding of the characteristics defining globally popular music on the platform.
Columns
- id: A unique identifier assigned to each song on Spotify.
- name: The title of the song.
- duration: The length of the song, measured in minutes.
- energy: A perceptual measure of intensity and activity, ranging from 0.0 to 1.0. High energy tracks typically feel fast, loud, and noisy, reflecting attributes like dynamic range, perceived loudness, timbre, onset rate, and general entropy.
- key: Indicates the musical key of the track, represented by integers mapping to standard Pitch Class notation (e.g., 0 for C, 1 for C♯/D♭). A value of -1 means no key was detected.
- loudness: The overall loudness of the track in decibels (dB), averaged across its duration. This value helps compare the relative loudness of different tracks and typically ranges between -60 and 0 dB.
- mode: Denotes the modality of a track (major or minor), indicating the type of scale from which its melodic content is derived. Major is represented by 1 and minor by 0.
- speechiness: Detects the presence of spoken words. Values above 0.66 suggest entirely spoken-word tracks, values between 0.33 and 0.66 indicate tracks with both music and speech (like rap), and values below 0.33 most likely represent music or non-speech content.
- acousticness: A confidence measure (from 0.0 to 1.0) of whether the track is acoustic, with 1.0 indicating high confidence in its acoustic nature.
- instrumentalness: Predicts the likelihood of a track containing no vocals. "Ooh" and "aah" sounds are considered instrumental, while rap or spoken word tracks are vocal. Values closer to 1.0 suggest a higher probability of the track being instrumental, with values above 0.5 generally indicating instrumental tracks.
- liveness: (Statistical data available, but no detailed description provided in sources.)
- valence: (Statistical data available, but no detailed description provided in sources.)
- tempo: (Statistical data available, but no detailed description provided in sources.)
- danceability: (Statistical data available, but no detailed description provided in sources.)
Distribution
The dataset is provided in a tabular format, typically as a CSV file. It comprises 100 records and features 14 distinct columns. The file size is approximately 10.48 kB.
Usage
This dataset is ideal for a range of applications, including:
- Exploratory Data Analysis: Gaining insights into the characteristics of highly successful songs.
- Feature Extraction: Understanding the specific audio features that contribute to song popularity.
- Audio Classification: Developing models to categorise or predict music genres or styles based on their features.
- Analysing trends in global music streaming.
Coverage
The dataset covers the most streamed songs of all time on Spotify globally. A related dataset focusing on the most streamed songs from 2021 is also available for comparative analysis.
License
CC0: Public Domain
Who Can Use It
This dataset is suitable for:
- Data Scientists: For building and testing machine learning models related to music.
- Music Analysts and Researchers: To study global music trends and song characteristics.
- Developers: For integrating music feature analysis into applications.
- Students: As a practical resource for data analysis and programming exercises.
Dataset Name Suggestions
- Spotify Global Top 100 Songs
- All-Time Most Streamed Spotify Tracks
- Spotify Audio Features Dataset
- Top 100 Streamed Songs Analysis
Attributes
Original Data Source: Spotify Audio Features Dataset