Analyze speakers | sync. labs

curl -X POST https://api.sync.so/v2/transcriptions/id/speaker-analysis \
     -H "x-api-key: <apiKey>" \
     -H "Content-Type: application/json" \
     -d '{
  "revision": "revision",
  "sourceVideoUrl": "sourceVideoUrl"
}'

{
  "revision": "revision",
  "sourceStartMs": 1,
  "status": "processing",
  "speakers": [
    {
      "id": "id",
      "status": "matched",
      "reason": "reason",
      "avatar": {
        "sourceTimeMs": 1.1,
        "box": {
          "x": 1.1,
          "y": 1.1,
          "width": 1.1,
          "height": 1.1
        },
        "sourceWidth": 1,
        "sourceHeight": 1
      },
      "faceTrack": {
        "url": "url",
        "fps": 1.1,
        "frameCount": 1,
        "sourceWidth": 1,
        "sourceHeight": 1
      },
      "evidence": {
        "supportMs": 1.1,
        "margin": 1.1,
        "meanScore": 1.1,
        "coverage": 1.1
      },
      "faceSupport": [
        {
          "identity": "identity",
          "supportMs": 1.1,
          "intervals": [
            {
              "startMs": 1.1,
              "endMs": 1.1
            },
            {
              "startMs": 1.1,
              "endMs": 1.1
            }
          ],
          "meanScore": 1.1
        },
        {
          "identity": "identity",
          "supportMs": 1.1,
          "intervals": [
            {
              "startMs": 1.1,
              "endMs": 1.1
            },
            {
              "startMs": 1.1,
              "endMs": 1.1
            }
          ],
          "meanScore": 1.1
        }
      ]
    },
    {
      "id": "id",
      "status": "matched",
      "reason": "reason",
      "avatar": {
        "sourceTimeMs": 1.1,
        "box": {
          "x": 1.1,
          "y": 1.1,
          "width": 1.1,
          "height": 1.1
        },
        "sourceWidth": 1,
        "sourceHeight": 1
      },
      "faceTrack": {
        "url": "url",
        "fps": 1.1,
        "frameCount": 1,
        "sourceWidth": 1,
        "sourceHeight": 1
      },
      "evidence": {
        "supportMs": 1.1,
        "margin": 1.1,
        "meanScore": 1.1,
        "coverage": 1.1
      },
      "faceSupport": [
        {
          "identity": "identity",
          "supportMs": 1.1,
          "intervals": [
            {
              "startMs": 1.1,
              "endMs": 1.1
            },
            {
              "startMs": 1.1,
              "endMs": 1.1
            }
          ],
          "meanScore": 1.1
        },
        {
          "identity": "identity",
          "supportMs": 1.1,
          "intervals": [
            {
              "startMs": 1.1,
              "endMs": 1.1
            },
            {
              "startMs": 1.1,
              "endMs": 1.1
            }
          ],
          "meanScore": 1.1
        }
      ]
    }
  ],
  "sourceVideoUrl": "sourceVideoUrl",
  "sourcePlaybackUrl": "sourcePlaybackUrl",
  "sourceEndMs": 1,
  "reason": "reason",
  "model": "model"
}

Matches each transcript speaker to the face that talks during their words. Read GET /v2/transcriptions/{id}/speaker-analysis first, and call this to start an analysis that wasn’t run or to retry one that ended unavailable. Pass the job’s revision and sourceVideoUrl. Supports up to 2 minutes of video and 8 speakers, and 3 requests per minute.

Authentification

x-api-keystring

Authentification par clé API via en-tête

Paramètres de chemin

idstringRequis

Transcription job id.

Requête

This endpoint expects an object.

revisionstringRequis

The revision of the transcription job.

sourceVideoUrlstringRequis

The transcription's source video URL.

retrybooleanOptionnel

Set to true to rerun an analysis that ended unavailable. A completed or running analysis is returned as is.

Réponse

Speaker analysis started or returned

revisionstring

The transcription revision this analysis belongs to.

sourceStartMsinteger

Start of the analyzed window, in source milliseconds.

statusenum

Status of the speaker analysis. processing is not terminal; completed and unavailable are.

Valeurs autorisées:

speakerslist of objects

One entry per transcript speaker, up to 8.

sourceVideoUrlstringOptionnel

The analyzed source video. Null for jobs created from sourceAudioUrl.

sourcePlaybackUrlstringOptionnel

A temporary URL for playing the source video.

sourceEndMsintegerOptionnel

End of the analyzed window, in source milliseconds. Null when the window runs to the end of the source.

reasonstringOptionnel

Why the analysis is unavailable or wasn't run.

modelstringOptionnel

The model that ran the analysis.

Erreurs

401

Unauthorized Error

404

Not Found Error

409

Conflict Error

422

Unprocessable Entity Error

429

Speaker Analysis Too Many Requests Error

500

Internal Server Error