Sign in

v1

latestOpenAPI 3.0.3RingCentral API License Agreement2026-08-064871,2921.5 MB
Audio

Speaker Diarization

Identifies who said what. Speaker diarization will identify the speaker for each segment so you can tell who spoke the sentence, paragraph, or phrase.

post/ai/audio/v1/async/speaker-diarize

Query parameters

webhookstring uri required

The webhook URI to which the job response will be returned

Request body

contentUristring uri

Publicly facing uri

encoding'Mpeg' | 'Mp4' | 'Wav' | 'Webm' | 'Webp' | 'Aac' | 'Avi' | 'Ogg' required

The encoding of the original audio

languageCodestring required

Language spoken in the audio file.

sourcestring

Source of the audio file eg: Phone, RingCentral, GoogleMeet, Zoom etc

audioType'CallCenter' | 'Meeting' | 'EarningsCalls' | 'Interview' | 'PressConference' | 'Voicemail'

Type of the audio

separateSpeakerPerChannelboolean

Set to True if the input audio is multi-channel and each channel has a separate speaker.

speakerCountinteger

Number of speakers in the file, omit parameter if unknown

speakerIdsstring[]

Optional set of speakers to be identified from the call.

enableVoiceActivityDetectionboolean

Apply voice activity detection.

Example request

{
  "encoding": "Wav",
  "languageCode": "en-US",
  "source": "RingCentral",
  "audioType": "CallCenter",
  "speakerCount": 2,
  "speakerIds": [
    "speakerId1",
    "speakerId2"
  ]
}

Response

Task accepted for processing.

jobIdstring