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.

versionVersionrequired

Constant value: VoiceoverV2

Default value: VoiceoverV2

voiceoverDegreeVoiceover Degreerequired

Possible values: >= 0 and <= 1

isVoiceoverDominantIs Voiceover Dominantrequired

segments objectrequired

Segment-level voiceover degree (0.0-1.0).

timestampsSecondsnumber[]required

Possible values: >= 0

valuesnumber[]required

Possible values: >= 0 and <= 1

VoiceoverV2

{
  "version": "VoiceoverV2",
  "voiceoverDegree": 0,
  "isVoiceoverDominant": true,
  "segments": {
    "timestampsSeconds": [
      0
    ],
    "values": [
      0
    ]
  }
}