Media API

Name: Media API
Brand: Capital Enesti

Turn YouTube videos and audio files into word-level transcripts, speaker diarization, voice fingerprints, and rich metadata with fast, reliable APIs.

Media$0.00+v1.2.1No reviewsOperational

Per request$0.00+up to $2.00 + proxy fee

Uptime100%Last 30d

Avg response195msLast 30d · <1k requests

API base URL

OpenAPI spec URL

Endpoints · 13 of 13

GET /health

curl -X GET https://eu-north-1.apimercado.com/v1/e72bpbz582mh/media-api/health \
  -H "X-API-Key: $API_KEY"

Request

{}

Response

{
  "data": {
    "db": "ok",
    "cache": {
      "exists": true,
      "writable": true
    },
    "status": "healthy",
    "configWarnings": [
      "string"
    ]
  },
  "meta": {
    "<key>": null
  },
  "error": null,
  "status": "ok"
}

About

Audio transcription, speaker diarization, voice embeddings, and metadata extraction for YouTube videos and for your own audio or video files.

Workflow

Asynchronous jobs: Transcripts, voice fingerprints, and audio embeddings run as background jobs. Submit a request and poll the endpoint (returns 202 Accepted while processing, and 200 OK once finished).
Synchronous endpoints: Video metadata and playlist endpoints return immediately.
Deterministic & Cached: Processed audio assets and metadata are cached immutably. Repeated requests return cached data immediately.

Files

To transcribe a file you store yourself (R2 or S3), register a presigned GET URL with POST /videos and use the returned vid_… id as videoId on POST /transcripts (engine deepgram only), POST /fingerprints and POST /embeddings. Only the URL is stored and the audio is fetched when the first job needs it, so keep the URL valid until then; an expired URL fails the job. Files up to 3 hours. DELETE /videos/{videoId} removes the Video and everything made from it.

Authentication

Authenticate all requests by including your API token in the Authorization header: Authorization: Bearer <token>

Responses

All responses use a standard envelope: { status, data, meta, error }.