Skip to main content

What are Music Audio Features?

Musical attributes based on a song’s composition, style, and mood.

Edouard avatar
Written by Edouard
Updated this week

Audio features are musical attributes extracted from streaming platforms to help analyze a song’s composition, style, and mood. These values are derived directly from platform-level analysis and are available through the Soundcharts API.

Below is a breakdown of each audio feature and what it represents:

Acousticness

A confidence score (0.0 to 1.0) estimating whether a track is acoustic. A value close to 1.0 means the system is highly confident the track is primarily acoustic.

Danceability

A measure (0.0 to 1.0) indicating how suitable a track is for dancing. It reflects tempo, beat stability, rhythm strength, and overall predictability. 0.0 = not danceable, 1.0 = highly danceable.

Energy

A perceptual score (0.0 to 1.0) representing intensity and activity. Energetic tracks usually sound fast, loud, and dense. For reference: death metal = high energy; a quiet classical prelude = low energy.

Instrumentalness

Estimates how likely a track is to contain no vocals. Values closer to 1.0 indicate instrumental content. Above 0.5, a track is likely instrumental; confidence increases as the value approaches 1.0.

Key

The musical key of the track encoded as an integer:

  • 0 = C

  • 1 = C♯/D♭

  • 2 = D

… up to 11

If no key is detected, the value is -1.

Liveness

Estimates the presence of an audience in the recording. Higher values increase the likelihood that the track was recorded live. Values above 0.8 strongly suggest a live performance.

Loudness

The overall loudness of the track measured in decibels (dB), averaged across the whole recording. Values typically range from -60 dB (quiet) to 0 dB (very loud). Helpful in comparing relative loudness between tracks.

Mode

Indicates whether the track is in:

  • 1 = Major

  • 0 = Minor

Speechiness

Measures the presence of spoken words.

  • > 0.66: mostly or entirely spoken (e.g., podcasts, audiobooks).

  • 0.33 – 0.66: mix of speech and music (e.g., rap, spoken segments).

  • < 0.33: primarily musical with little to no speech.

Tempo

Estimated tempo in beats per minute (BPM). It reflects how fast or slow the track feels.

Time Signature

Estimated number of beats per bar. Values range from 3 to 7, representing meters like 3/4 through 7/4.

Example: 4 = traditional 4/4 time.

Valence

A score (0.0 to 1.0) describing the emotional “positiveness” of the track.

High valence = happy, bright, euphoric.

Low valence = sad, tense, or darker mood.

Did this answer your question?