Model Outputs

A model is an entity of Cyanite's AI that takes an audio track as input and produces a specific type of musical information as output (for example: mood scores, BPM, or a key prediction).

A model output is the versioned JSON object returned by one such model.

tip

New to Cyanite? Start with the Quick Start Guide to upload your first track. Then use the Get Models endpoint to retrieve model outputs.

This page documents the public model outputs currently exposed to API consumers. It focuses on field structure and meaning.

For taxonomy-based outputs (for example: moods, genres, instruments), the possible taxonomy values are listed in the Appendix: Taxonomy vocabularies for reference.

---

Example model output

The following example shows a taxonomy-based model output (MoodSimpleV2). It demonstrates the common scores / tags / segments pattern.

The example is explained in the sections below (see Taxonomy classifiers and Track-level vs segment-level data).

{
  "version": "MoodSimpleV2",
  "segments": {
    "timestampsSeconds": [5.0, 15.0, 25.0],
    "values": {
      "aggressive": [0.2172, 0.2986, 0.2332],
      "calm": [0.216, 0.1058, 0.2894],
      "chill": [0.1954, 0.0933, 0.0839],
      "dark": [0.4876, 0.2672, 0.5492],
      "energetic": [0.5873, 0.635, 0.2871],
      "epic": [0.0484, 0.2499, 0.0261],
      "happy": [0.2862, 0.136, 0.0894],
      "romantic": [0.0218, 0.0706, 0.0022],
      "sad": [0.2462, 0.0899, 0.3842],
      "scary": [0.3158, 0.0813, 0.2322],
      "sexy": [0.3539, 0.3665, 0.1121],
      "ethereal": [0.224, 0.1265, 0.2627],
      "uplifting": [0.3555, 0.428, 0.1668]
    }
  },
  "scores": {
    "aggressive": 0.2497,
    "calm": 0.2037,
    "chill": 0.1242,
    "dark": 0.4347,
    "energetic": 0.5032,
    "epic": 0.1082,
    "happy": 0.1705,
    "romantic": 0.0315,
    "sad": 0.2401,
    "scary": 0.2098,
    "sexy": 0.2775,
    "ethereal": 0.2044,
    "uplifting": 0.3168
  },
  "tags": ["energetic", "dark"]
}

Versioning

Every model output object contains a version field.

The version is a discriminator (so you can pick the right schema), and it also tracks breaking changes to the output structure. It defines:

• Which schema to expect (fields and types)
• Which taxonomy/vocabulary applies (for taxonomy-based outputs)

Multiple versions may exist at the same time. Older versions are deprecated over time and API consumers should migrate to the latest available version when possible.

---

Track-level vs segment-level data

Many model outputs provide both:

• Track-level results: a single prediction for the whole track (e.g. tag, scores, tags).
• Segment-level results: predictions over time under segments.

Practical guidance:

• If you need one value or descriptor for the track as a whole, use the track-level fields (tag, scores, tags).
• If you need how values evolve over the course of the track (for example: where a mood becomes more intense), use segments.

Where present, segment-level data is represented as:

• segments.timestampsSeconds: array of segment start times (seconds from track start)
• segments.values: the segment values aligned by index with timestampsSeconds

The shape of segments.values depends on the model output:

• Taxonomy classifiers: segments.values is an object mapping taxonomy keys to arrays of scores.
• Musical parameters: segments.values is an array of predicted values.

In both cases, the array length(s) match segments.timestampsSeconds.length.

---

Common field patterns

Taxonomy classifiers (scores, segments, tags)

Many classifiers share the same structure:

• scores: object mapping taxonomy keys → track-level score (usually 0.0-1.0), indicating the model's predicted probability for each tag.
• segments: segment-level scores over time.
• tags: array of the dominant taxonomy keys (a subset of the keys in scores).

For these model outputs, the precise set of taxonomy keys is fixed for the given version.

Musical parameters (tag, segments, confidence)

Musical parameter model outputs (e.g. BPM, key, time signature) expose:

• tag: the predicted global value for the whole track.

• segments: the predicted local values per segment.

• confidence: a structured confidence object with three values in 0.0-1.0:

  • modelCertainty: how sure the model is that the predictions are correct
  • predictionStability: how stable the predictions are across segments
  • confidence: an overall confidence score, the minimum of the two individual scores

  Interpretation: Confidence values above 0.9 indicate the track is unambiguous and a single global label can be confidently applied.

---

List of model outputs

The following model output versions are public (alphabetical):

• AiMusicDetectionV1
• AudioFileInfoV1
• AugmentedKeywordsV3
• AutoDescriptionV2
• BpmV2
• CharacterV2
• FreeGenreV3
• InstrumentsV2
• KeyV2
• MainGenreV2
• MoodAdvancedV2
• MoodSimpleV2
• MovementV2
• MusicForV1
• MusicalEraV2
• RepresentativeSegmentV2
• SegmentationV1
• SubgenreV2
• TempoV1
• TimeSignatureV2
• ValenceArousalV2
• VocalsV2
• VocalStyleV1
• VoiceoverV2

---

AiMusicDetectionV1

Detects whether a track is likely AI-generated music.

Use this model to power UI scanning, filtering, or downstream policy decisions where you want an estimate of whether a track was generated by an AI music model.

Taxonomy: none

Field	Type	Description
version	string	Always "AiMusicDetectionV1".
isAiMusic	boolean	true if the track is predicted as AI-generated music.
score	number	Confidence-like score in 0.0-1.0. Higher means more likely AI-generated.
suspectedModel	string | null	Optional information about which generative model is suspected (if predicted as AI-generated).

AudioFileInfoV1

Technical properties of the provided audio file.

Taxonomy: none

Field	Type	Description
version	string	Always "AudioFileInfoV1".
duration	number	Track duration in seconds.
fileSizeB	integer	File size in bytes.
bitrate	integer	Bitrate value as reported by decoding/metadata.
samplerate	integer	Sample rate in Hz.

AugmentedKeywordsV3

Keyword-like descriptors inferred from audio, represented as a weighted dictionary.

Taxonomy: none

Field	Type	Description
version	string	Always "AugmentedKeywordsV3".
scores	object	Map of 20 highest scoring keyword identifiers → score in -1.0-1.0 (higher means stronger association).

AutoDescriptionV2

A short natural-language description of the track.

This output is designed for human-facing UI (e.g. previews, catalog cards). Treat it as generated text that may vary by model version.

Taxonomy: none

Field	Type	Description
version	string	Always "AutoDescriptionV2".
description	string	Text description generated for the track.

BpmV2

Tempo in beats per minute (BPM).

Taxonomy: none

Field	Type	Description
version	string	Always "BpmV2".
tag	integer	Track-level BPM estimate (constrained to 60-200).
segments	object	Segment-level BPM values.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	integer[]	BPM per segment (aligned with timestampsSeconds, 60-200).
confidence	object	Musical-parameter confidence object (all values 0.0-1.0).
confidence.modelCertainty	number	How certain the model is in its local predictions.
confidence.predictionStability	number	How stable the BPM is over time.
confidence.confidence	number	Overall confidence score. Values above 0.9 indicate the track is unambiguous and a single global label applies.

CharacterV2

High-level “character” descriptors (taxonomy-based classifier).

Use this model to describe the overall sonic “personality” of a track (e.g. for playlisting, similarity, and creative search).

Taxonomy: CharacterV2Tags

Field	Type	Description
version	string	Always "CharacterV2".
scores	object	Track-level scores (0.0-1.0) keyed by character tags.
tags	string[]	Dominant character tags (subset of keys in scores).
segments	object	Segment-level scores.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	object	Map of character tags → number[] (aligned with timestampsSeconds).

FreeGenreV3

Free-form genre tags without a fixed taxonomy.

Use this model when you want flexible genre labeling, including niche genres not covered by a fixed taxonomy. Unlike taxonomy-based genre models, this output does not provide per-tag scores.

Taxonomy: none

Field	Type	Description
version	string	Always "FreeGenreV3".
tags	string[]	Free-text / free-vocabulary genre tags.

InstrumentsV2

Instrument presence classifier (taxonomy-based).

This model can be used to detect which instruments are present and how strongly, and to localize those instruments over time via segments.

Taxonomy: InstrumentsV2Tags (instruments), InstrumentsV2PresenceTags (presence categories)

Field	Type	Description
version	string	Always "InstrumentsV2".
presence	object	Track-level instrument presence categories keyed by instrument identifier.
tags	string[]	Detected instrument tags (subset of instrument taxonomy), sorted by amount of occurence
segments	object	Segment-level instrument scores.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	object	Map of instrument identifier → number[] scores (aligned with timestampsSeconds).

KeyV2

Musical key prediction.

Taxonomy: KeyV2Tags

Field	Type	Description
version	string	Always "KeyV2".
tag	string	Track-level key detection from the KeyV2 taxonomy.
segments	object	Segment-level key tags.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	string[]	Key per segment (aligned with timestampsSeconds).
confidence	object	Musical-parameter confidence object (all values 0.0-1.0).
confidence.modelCertainty	number	Overall certainty of the key model.
confidence.predictionStability	number	How stable the key is over time.
confidence.confidence	number	Overall confidence score. Values above 0.9 indicate the track is unambiguous and a single global label applies.

MainGenreV2

Main genre classifier (taxonomy-based).

Use this model for high-level genre filtering and organization. For a more specific genre classification, see SubgenreV2 and FreeGenreV3.

Taxonomy: MainGenreV2Tags

Field	Type	Description
version	string	Always "MainGenreV2".
scores	object	Track-level scores (0.0-1.0) keyed by genre tags.
tags	string[]	Dominant genre tags (subset of keys in scores).
segments	object	Segment-level scores.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	object	Map of genre → number[] scores (aligned with timestampsSeconds).

MoodAdvancedV2

Advanced mood classifier (taxonomy-based).

Use this model when you want a more fine-grained emotional description of the track. If you prefer a smaller, simpler mood vocabulary, use MoodSimpleV2.

Taxonomy: MoodAdvancedV2Tags

Field	Type	Description
version	string	Always "MoodAdvancedV2".
scores	object	Track-level scores (0.0-1.0) keyed by mood tags.
tags	string[]	Dominant mood tags (subset of keys in scores).
segments	object	Segment-level scores.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	object	Map of mood tags → number[] scores (aligned with timestampsSeconds).

MoodSimpleV2

Simple mood classifier (taxonomy-based).

Use this model for a compact mood description that is easy to surface in UX and filtering. For more granular moods, use MoodAdvancedV2.

Taxonomy: MoodSimpleV2Tags

Field	Type	Description
version	string	Always "MoodSimpleV2".
scores	object	Track-level scores (0.0-1.0) keyed by mood tags.
tags	string[]	Dominant mood tags (subset of keys in scores).
segments	object	Segment-level scores.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	object	Map of mood tags → number[] scores (aligned with timestampsSeconds).

MovementV2

Rhythmic “movement” classifier (taxonomy-based).

Use this model to describe rhythmic feel (groove / motion) beyond raw BPM. This is often helpful for search and playlisting based on “feel”.

Taxonomy: MovementV2Tags

Field	Type	Description
version	string	Always "MovementV2".
scores	object	Track-level scores (0.0-1.0) keyed by movement tags.
tags	string[]	Dominant movement tags (subset of keys in scores).
segments	object	Segment-level scores.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	object	Map of movement tags → number[] scores (aligned with timestampsSeconds).

MusicForV1

Use-case oriented tags indicating what the track is suitable for.

Use this model for “suitability” / “music for …” style metadata (e.g. mapping tracks to use-cases in product UX or search).

Taxonomy: MusicForV1Tags

Field	Type	Description
version	string	Always "MusicForV1".
tags	string[]	Dominant music-for tags.

MusicalEraV2

Estimated production year and musical-era label.

Use this model to describe when a track sounds like it was produced (helpful for era filtering and catalog metadata).

Taxonomy: MusicalEraV2Tags

Field	Type	Description
version	string	Always "MusicalEraV2".
estimatedProductionYear	integer	Estimated year, from 1950
tag	string	Era tag from the MusicalEraV2 taxonomy.

RepresentativeSegmentV2

The most representative segment of the track.

Use this model to pick a short segment that best represents the track overall (e.g. for previews or highlighting). Taxonomy: none

Field	Type	Description
version	string	Always "RepresentativeSegmentV2".
startSeconds	number	Start time (seconds) of the representative segment.
endSeconds	number	End time (seconds) of the representative segment.

SegmentationV1

Structural segmentation of the track into consecutive, non-overlapping sections along the audio timeline.

Each segment identifies a section by its start and end time in seconds. These segments are ordered sequentially so that one section ends exactly where the next begins.

Use this model to navigate a track's structure, target search boundaries, drive waveform/section UIs, cueing, or pick section boundaries for previews and edits.

Taxonomy: none

Field	Type	Description
version	string	Always "SegmentationV1".
segments	object[]	Ordered, non-overlapping sections identified within the track timeline.
segments.start	number	Segment start time in seconds from the beginning of the track.
segments.end	number	Segment end time in seconds (matches the start of the next segment).

SubgenreV2

Subgenre classifier.

Use this model when you need a more specific classification than MainGenreV2. If the track does not fit the subgenre taxonomy with sufficient confidence, the output is null.

Subgenres are only detected within their corresponding parent-genre context; if the track does not match any subgenre under that hierarchy, this output will be null.

Taxonomy: SubgenreV2Tags

This model output can be null/missing (represented as null values for scores, tags, and segments) when no subgenre can be determined with sufficient confidence.

If segments is present, individual entries inside segments.values can still be null.

Field	Type	Description
version	string	Always "SubgenreV2".
scores	object | null	Track-level scores (0.0-1.0) keyed by subgenre tags.
tags	string[] | null	Dominant subgenre tags.
segments	object | null	Segment-level scores.
segments.timestampsSeconds	number[]	Segment start times in seconds (only present when segments is not null).
segments.values	object	Map of subgenre → number[] | null scores (aligned with timestampsSeconds).

TempoV1

High-level "perceived" tempo classification (taxonomy-based), plus a track-level score.

Use this model when you want a coarse tempo label (fast/slow-style categories) rather than a numeric BPM. For numeric tempo, use BpmV2.

Taxonomy: TempoV1Tags

Field	Type	Description
version	string	Always "TempoV1".
tag	string	Tempo tag from the TempoV1 taxonomy.
score	number	Track-level score in 0.0-1.0 for the predicted tag.
segments	object	Segment-level tempo scores for the predicted tempo tag.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	number[]	Score per segment (0.0-1.0, aligned with timestampsSeconds).

TimeSignatureV2

Time signature prediction.

Use this model to determine the track’s meter (e.g. 4/4, 3/4).

Taxonomy: none

Field	Type	Description
version	string	Always "TimeSignatureV2".
tag	string	Track-level time signature (e.g. "4/4").
segments	object	Segment-level time signature values.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	string[]	Time signature per segment (aligned with timestampsSeconds).
confidence	object	Musical-parameter confidence object (all values 0.0-1.0).
confidence.modelCertainty	number	Overall certainty of the time signature model.
confidence.predictionStability	number	How stable the time signature is over time.
confidence.confidence	number	Overall confidence score. Values above 0.9 indicate the track is unambiguous and a single global label applies.

ValenceArousalV2

Continuous emotional axes and derived categorical descriptors.

Use this model for continuous emotional metrics (valence/arousal) and for derived, user-friendly categorical summaries such as energy and emotion-change labels.

Taxonomy: ValenceArousalV2Tags (axes), ValenceArousalV2EnergyLevelTags, ValenceArousalV2EnergyChangesTags, ValenceArousalV2EmotionProfileTags, ValenceArousalV2EmotionChangesTags

Field	Type	Description
version	string	Always "ValenceArousalV2".
scores	object	Track-level axis values (constrained to -1.0-1.0).
segments	object	Segment-level axis values.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	object	Map of axis key → number[] values (aligned with timestampsSeconds).
energyLevel	string	Overall energy level category.
energyChanges	string	How much energy changes over time.
emotionProfile	string	Overall emotional profile category.
emotionChanges	string	How much emotion changes over time.

VocalsV2

Vocal-characteristic classifier (taxonomy-based), with optional summary fields.

Use this model to understand whether and how much vocals are present and what gender they resemble.

Taxonomy: VocalsV2Tags, VocalsV2PresenceTags, VocalsV2PredominantVocalGenderTags

Field	Type	Description
version	string	Always "VocalsV2".
scores	object	Track-level scores (0.0-1.0) keyed by vocal tags.
tags	string[]	Dominant vocal tags.
segments	object	Segment-level scores.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	object	Map of vocal tags → number[] scores (aligned with timestampsSeconds).
vocalPresence	string | null	Optional overall vocal presence label (if available).
predominantVocalGender	string | null	Optional predominant vocal gender label (if available).

VocalStyleV1

Vocal style classifier (taxonomy-based).

Use this model to describe the characteristics of present vocals in more detail (useful for more specific vocal characterization beyond general vocals presence and gender).

Taxonomy: VocalStyleV1Tags

Field	Type	Description
version	string	Always "VocalStyleV1".
scores	object	Track-level scores (0.0-1.0) keyed by vocal-style tags.
tags	string[]	Dominant vocal-style tags.
segments	object	Segment-level scores.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	object	Map of vocal-style tags → number[] scores (aligned with timestampsSeconds).

VoiceoverV2

Voiceover detection and degree estimation.

Use this model to detect spoken-word / talkover content and to estimate how dominant voiceover is across the track.

Taxonomy: none

Field	Type	Description
version	string	Always "VoiceoverV2".
voiceoverDegree	number	Degree of voiceover dominance in 0.0-1.0.
isVoiceoverDominant	boolean	true if voiceover is dominant.
segments	object	Segment-level voiceover degree.
segments.timestampsSeconds	number[]	Segment start times in seconds.
segments.values	number[]	Voiceover degree per segment (0.0-1.0, aligned with timestampsSeconds).

---

Appendix: Taxonomy vocabularies

The following taxonomies are referenced by the public model outputs above. They are listed here as plain vocabularies (no explanations).

CharacterV2Tags

Values

bold
cool
epic
ethereal
heroic
luxurious
magical
mysterious
playful
powerful
retro
sophisticated
sparkling
sparse
unpolished
warm

InstrumentsV2Tags

Values

accordion
acousticGuitar
africanPercussion
asianFlute
asianStrings
banjo
bass
bassGuitar
bells
bongoConga
brass
celeste
cello
churchOrgan
clarinet
doubleBass
drumKit
electricGuitar
electricOrgan
electricPiano
electronicDrums
flute
horn
glockenspiel
harp
harpsichord
luteOud
mandolin
marimba
oboe
percussion
piano
pizzicato
saxophone
sitar
steelDrums
strings
synth
tabla
taiko
trumpet
tuba
ukulele
vibraphone
violin
whistling
woodwinds

InstrumentsV2PresenceTags

Values

absent
partially
frequently
throughout

KeyV2Tags

Values

aMajor
aMinor
bMajor
bMinor
bFlatMajor
bFlatMinor
cMajor
cMinor
dFlatMajor
cSharpMinor
dMajor
dMinor
eFlatMajor
dSharpMinor
eMajor
eMinor
fMajor
fMinor
fSharpMajor
fSharpMinor
gMajor
gMinor
aFlatMajor
gSharpMinor

MainGenreV2Tags

Values

african
ambient
middleEastern
asian
blues
childrenJingle
classical
electronic
folkCountry
funkSoul
indian
jazz
latin
metal
pop
rapHipHop
reggae
rnb
rock
singerSongwriter
sound
soundtrack
spokenWord

MoodAdvancedV2Tags

Values

adventurous
aggressive
agitated
angry
anthemic
anxious
aweInspiring
barren
bittersweet
blue
boingy
boisterous
bright
calm
celebratory
cheerful
cold
comedic
concerned
confident
contented
cool
courageous
creepy
dangerous
dark
delicate
depressing
determined
dignified
disturbing
dreamy
eccentric
eerie
emotional
energetic
epic
euphoric
evil
excited
exciting
exhilarating
fearful
feelGood
fiery
frightening
fun
funny
gloomy
graceful
happy
heavy
heroic
hopeful
horror
innocent
inspirational
intense
intimate
introspective
joyous
kind
laidBack
leisurely
light
lighthearted
lonely
loving
lyrical
majestic
melancholic
menacing
mischievous
motivational
mournful
mysterious
nervous
nightmarish
noble
nostalgic
ominous
optimistic
panicStricken
passionate
peaceful
perky
playful
poignant
ponderous
positive
powerful
prestigious
proud
quiet
quirky
reflective
relaxed
relentless
resolute
restless
rollicking
romantic
sad
scary
seductive
sentimental
serene
serious
sexy
soaring
soft
solemn
soothing
spiritual
spooky
strange
strong
supernatural
suspenseful
sweet
tender
tense
thoughtful
tranquil
triumphant
upbeat
uplifting
victorious
violent
warm
weird
whimsical

MoodSimpleV2Tags

Values

aggressive
calm
chill
dark
energetic
epic
happy
romantic
sad
scary
sexy
ethereal
uplifting

MovementV2Tags

Values

bouncing
driving
flowing
groovy
nonrhythmic
pulsing
robotic
running
steady
stomping

MusicForV1Tags

Values

achievement
action
adventure
advertising
alien
animation
animals
anniversary
anthem
anticipation
artInstallation
atmosphere
baby
backdrop
background
bar
barbecue
battle
beach
bedroom
branding
breakup
broadcast
building
bumper
business
cabaret
cafe
campaign
campfire
car
carChase
carol
cartoon
celebration
ceremony
chase
cheer
children
chillOut
choir
christmas
church
cinema
cinematic
circus
city
climax
closingCredits
club
coffeeShop
college
comedy
comingOfAge
commerce
commercial
community
competition
confidence
conflict
cooking
corporate
countryside
creativity
crime
crowd
cypher
dance
dancing
danger
darkness
dateNight
death
delight
desert
destruction
detective
dinner
discovery
djSet
documentary
drama
dream
drinks
drive
driving
education
elevator
empowerment
energy
engagement
enjoying
ensemble
entertainment
era
espionage
evening
event
exercise
explanation
exploration
fairyTale
faith
family
fantasy
farewell
farm
fashion
fashionShow
fear
festival
fight
film
fitness
flight
focus
freedom
friendship
funeral
future
gallery
gameTrailer
gaming
ghost
glory
goodTimes
grinding
gym
halloween
hero
highlights
historical
history
holiday
hollywood
home
horror
hotel
hunt
hype
immersion
improvisation
incidental
indie
infographic
infomercial
inspiration
installation
instrumental
intimacy
intrigue
intro
introspection
investigation
island
jingle
journey
jungle
karaoke
lecture
leisure
lifestyle
lobby
lookbook
loss
lounge
love
lullaby
luxury
meditation
memorial
memories
military
monster
montage
morning
motivation
motorsport
movie
museum
mystery
narration
narrative
nature
newYear
news
night
nightDrive
nightclub
nightlife
nostalgia
ocean
orchestra
outdoor
outro
parade
party
patriotism
performance
periodDrama
periodPiece
phobia
play
pleasure
podcast
presentation
promo
promotion
protest
pursuit
racing
radio
rain
rainyDay
rally
rave
reading
rebellion
recital
reel
rehearsal
relationship
relaxation
religion
restaurant
retro
roadTrip
robot
romance
sales
school
sciFi
science
sea
searching
shopping
show
singAlong
skateboarding
sleep
slideshow
smile
snow
soundDesign
soundEffects
soundscape
soundtrack
spa
space
sports
spring
spy
stinger
store
story
storytelling
street
study
summer
sunset
sunshine
superhero
suspense
teamwork
teaser
teen
television
theatre
thinking
thriller
timeLapse
title
trailer
training
transition
travel
tribute
triumph
tutorial
underscore
urban
vacation
videoGame
victory
vintage
vinyl
vlog
walk
walking
war
wedding
wellness
western
wildlife
winter
women
work
workout
worship
youth
youtube
zombie

MusicalEraV2Tags

Values

earlyMid1950s
midLate1950s
late1950sEarly1960s
earlyMid1960s
midLate1960s
late1960sEarly1970s
earlyMid1970s
midLate1970s
late1970sEarly1980s
earlyMid1980s
midLate1980s
late1980sEarly1990s
earlyMid1990s
midLate1990s
late1990sEarly2000s
earlyMid2000s
midLate2000s
late2000sEarly2010s
contemporary

SubgenreV2Tags

Values

abstractIdm
breakbeatDnb
deepHouse
electro
house
minimal
synthPop
techHouse
techno
trance
medievalPeriod
renaissancePeriod
baroquePeriod
classicalPeriod
romanticPeriod
contemporaryClassical
bluesRock
folkRock
hardRock
indieAlternative
psychProgRock
punk
rockAndRoll
popSoftRock
contemporaryRnb
gangsta
jazzyHipHop
popRap
trap
blackMetal
deathMetal
doomMetal
heavyMetal
metalcore
nuMetal
disco
funk
gospel
neoSoul
soul
bigBandSwing
bebop
contemporaryJazz
easyListening
fusion
latinJazz
smoothJazz
country
folk

TempoV1Tags

Values

slow
mediumSlow
medium
mediumFast
fast

ValenceArousalV2Tags

Values

valence
arousal

ValenceArousalV2EnergyLevelTags

Values

low
medium
high
varying

ValenceArousalV2EnergyChangesTags

Values

low
medium
high

ValenceArousalV2EmotionProfileTags

Values

negative
neutral
positive
varying

ValenceArousalV2EmotionChangesTags

Values

low
medium
high

VocalsV2Tags

Values

female
male
instrumental

VocalsV2PresenceTags

Values

low
medium
high

VocalsV2PredominantVocalGenderTags

Values

male
female

VocalStyleV1Tags

Values

femaleACapella
femaleChoir
femaleForegroundVocals
femaleBackgroundVocals
instrumental
maleACapella
maleChoir
maleForegroundVocals
maleBackgroundVocals
mixedACapella
mixedChoir
syntheticACapella
syntheticChoir
syntheticForegroundVocals
syntheticBackgroundVocals