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.

versionVersionrequired

Constant value: VocalsV2

Default value: VocalsV2

segments objectrequired

Segment-level scores for vocal tags.

timestampsSecondsnumber[]required

Possible values: >= 0

values objectrequired

Map of vocal tag → array of scores (aligned with timestampsSeconds).

femalenumber[]required

Possible values: >= 0 and <= 1

malenumber[]required

Possible values: >= 0 and <= 1

instrumentalnumber[]required

Possible values: >= 0 and <= 1

scores objectrequired

Track-level scores (0.0-1.0) keyed by vocal tag.

femaleFemalerequired

Possible values: >= 0 and <= 1

maleMalerequired

Possible values: >= 0 and <= 1

instrumentalInstrumentalrequired

Possible values: >= 0 and <= 1

tagsVocalsV2Tags[]required

Possible values: [female, male, instrumental]

vocalPresence object

anyOf

• VocalsV2PresenceTags
• null

Overall vocal presence label.

VocalsV2PresenceTags

Overall vocal presence label.

Possible values: [low, medium, high]

predominantVocalGender object

anyOf

• VocalsV2PredominantVocalGenderTags
• null

Predominant vocal gender label.

VocalsV2PredominantVocalGenderTags

Predominant vocal gender label.

Possible values: [male, female]

VocalsV2

{
  "version": "VocalsV2",
  "segments": {
    "timestampsSeconds": [
      0
    ],
    "values": {
      "female": [
        0
      ],
      "male": [
        0
      ],
      "instrumental": [
        0
      ]
    }
  },
  "scores": {
    "female": 0,
    "male": 0,
    "instrumental": 0
  },
  "tags": [
    "female"
  ],
  "vocalPresence": "low",
  "predominantVocalGender": "male"
}