Skip to main content

KeyV2

Musical key prediction. Provides track-level key detection and segment-level key tags over time, along with confidence scores.

versionVersionrequired
Constant value: KeyV2
Default value: KeyV2
segments objectrequired

Segment-level key tags.

timestampsSecondsnumber[]required

Start time of each segment in seconds from the beginning of the track.

Possible values: >= 0

valuesKeyV2Tags[]required

Key of each segment, aligned by index with timestampsSeconds.

Possible 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]

tagKeyV2Tagsrequired

Musical keys.

Possible 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]

confidences objectrequired

Confidence of the prediction, as three values between 0.0 and 1.0.

modelCertaintyModel Certaintyrequired

How sure the model is that the prediction is correct.

Possible values: >= 0 and <= 1

predictionStabilityPrediction Stabilityrequired

How stable the prediction is across segments.

Possible values: >= 0 and <= 1

confidenceConfidencerequired

Overall confidence, the minimum of modelCertainty and predictionStability. Values above 0.9 indicate the track is unambiguous and a single global label applies.

Possible values: >= 0 and <= 1

KeyV2
{
"version": "KeyV2",
"segments": {
"timestampsSeconds": [
0
],
"values": [
"aMajor"
]
},
"tag": "aMajor",
"confidences": {
"modelCertainty": 0,
"predictionStability": 0,
"confidence": 0
}
}