Enhanced Metadata
This page provides a high-level breakdown of enhanced metadata fields. Below, you will find an overview of the specific data fields included in the schema, including descriptions of how they are measured and the variety of classes they return.
Click HERE for a detailed overview of each field's classes.
Identifiers
Identification and versioning for the track.
| Name | Description |
|---|---|
| Version | An incremental integer representing the specific model version used to generate the enhanced metadata. |
| Release ID | The unique identifier of the release. |
| Track ID | The unique MassiveMusic identifier of the track. |
Embedding
The high-dimensional vector representation of the track.
| Name | Description | Type |
|---|---|---|
| Embedding | A 768-dimension vector representing the sonic essence of the track. More information about it can be found HERE. | Array |
Content
The type of the audio file, distinguishing between music and non-musical content.
| Name | Description | Class Count | Type (Range) |
|---|---|---|---|
| Content | Whether the track is music or not. | 2 | Distribution (0 to 1) |
Genres
The musical identity of the track, ranging from the broad "style" to granular sub-styles.
| Name | Description | Class Count | Type (Range) |
|---|---|---|---|
| Main Genre | The main genre(s) of the track. Up to 3 main genres returned. | 16 | Distribution (0 to 1) |
| Sub Genre | The sub genre(s) of the track. Up to 2 sub genres returned. | 135 | Distribution (0 to 1) |
Mood
The atmospheric "vibe" or personality of a track.
| Name | Description | Class Count | Type (Range) |
|---|---|---|---|
| Mood | The musical mood(s) of the track. Up to 3 moods returned. | 22 | Distribution (0 to 1) |
Emotion
The emotional response a listener is likely to have based on musical characteristics.
| Name | Description | Class Count | Type (Range) |
|---|---|---|---|
| Arousal | The level of physical activation evoked by the track. | 3 | Score (-1 to 1) |
| Valence | The level of "positivity" vs "negativity" evoked by the track. | 3 | Score (-1 to 1) |
| Engagement | The level of listener involvement or attention evoked by the track. | 3 | Score (-1 to 1) |
| Pleasantness | The level of enjoyment or comfort evoked by the track. | 3 | Score (-1 to 1 |
Instrumentation
The primary sound sources of a track, specifically focusing on voices and lead instruments.
| Name | Description | Class Count | Type (Range) |
|---|---|---|---|
| Vocals | The primary vocal type of the track. | 5 | Distribution (0 to 1) |
| Instruments | The most prominent instruments of a track. Up to 3 instruments returned. | 19 | Distribution (0 to 1) |
Sound
The technical and perceived sound characteristics of a track.
| Name | Description | Class Count | Type (Range) |
|---|---|---|---|
| Recording Environment | Whether a track was recorded in a studio or live environment. | 2 | Distribution (0 to 1) |
| Sound Generation | The dominant method of sound creation of a track. Acoustic describes a track with no electric/digital instruments, Electric describes a track with electric instruments such as an electric guitar, and Synthetic describes a track with digital/synth instruments such as a synthesizer. | 3 | Distribution (0 to 1) |
| Loudness | Measure in LKFS in accordance with the ITU-R BS.1770-4 standard. Note that LUFS is the same measurement as both measure loudness in the same way where 1 unit equals 1 decibel (1 dB). | 3 | Score (-100 to 0) |
| Roughness | The irregular amplitude fluctuations of a track. | 3 | Score (-1 to 1) |
| Timbre | The tonal quality or colour of the track. | 3 | Score (-1 to 1) |
| Space | The perceived spatial characteristics of the track. | 3 | Score (-1 to 1) |
| Texture | The overall density and layering of sounds in the track. | 3 | Score (-1 to 1) |
Harmonics
The tonal and melodic framework of the track.
| Name | Description | Class Count | Type (Range) |
|---|---|---|---|
| Harmony | The overall pleasantness of note combinations of a track. | 2 | Distribution (0 to 1) |
| Scale | The tonal framework of a track. | 2 | Distribution (0 to 1) |
| Key | The root note around which the track is centred. | 12 | Distribution (0 to 1) |
| Tonality | The overall melodic and tonal coherence of a track. | 4 | Distribution (0 to 1) |
Rhythmics
The rhythmic characteristics of a track, focusing on beat, pulse, and timing.
| Name | Description | Class Count | Type (Range) |
|---|---|---|---|
| Rhythm | The main timing pattern or time signature of a track. | 3 | Distribution (0 to 1) |
| Grooviness | The rhythmic feel of the track that creates the urge to move. | 3 | Score (-1 to 1) |
| Tempo | The speed or pace of the track. | 3 | Distribution (0 to 1) |
| BPM | The beats per minute of the track. | 1 | Integer |
Audience
Demographic insights into which groups are most likely to enjoy a track.
| Name | Description | Class Count | Type (Range) |
|---|---|---|---|
| Age | The audience age group(s) of a track. Up to 3 age groups returned. | 6 | Distribution (0 to 1) |
| Gender | The audience gender identity of a track. | 3 | Distribution (0 to 1) |
| Region | The audience geographical region(s) of a track. Up to 3 audience regions returned. | 12 | Distribution (0 to 1) |
Origin
Historical and geographic "roots" reflected in the track's style.
| Name | Description | Class Count | Type (Range) |
|---|---|---|---|
| Decade | The decade the track's musical and sound characteristics reflect the most. | 9 | Distribution (0 to 1) |
| Region | The geographic region a track's musical style are most closely associated with. | 12 | Distribution (0 to 1) |
Use Case
The best practical settings or activities for a track.
| Name | Description | Class Count | Type (Range) |
|---|---|---|---|
| Use Case | The most suitable contexts for a track's use. Up to 2 use cases returned. | 15 | Distribution (0 to 1) |
Updated 3 days ago