Transcribe a media file | Media API

Transcribes one video or audio URL and returns the text, per-segment and per-word timestamps, and the detected language. A creative with no speech returns an empty transcript with has_audio: false rather than an error.

5 credits / requestBilled only on success

Authorization

ApiKeyAuth

x-api-key<token>

In: header

Request Body

application/json

TypeScript Definitions

Use the request body type in TypeScript.

url*string

Public http(s) URL of one video or audio file. Ad-library creative URLs expire — fetch the ad and analyse it in the same run.

Formaturi

Lengthlength <= 4096

proxy_tier?string

Connection tier used to download the media file. none uses a direct connection.

Default"none"

Value in

"none"
"datacenter"
"residential"

language?string

Force a language instead of detecting it. Detection costs nothing extra, so set this only when you already know.

Match^[a-z]{2}$

curl -X POST "https://example.com/v1/media/transcribe" \  -H "x-api-key: YOUR_API_KEY" \  -H "Content-Type: application/json" \  -d '{    "url": "https://video.example.com/creative.mp4"  }'

{  "data": {    "language": "string",    "language_probability": 0,    "duration": 0,    "has_audio": true,    "text": "string",    "segments": [      {        "id": 0,        "start": 0,        "end": 0,        "text": "string",        "no_speech_prob": 0,        "avg_logprob": 0      }    ],    "words": [      {        "word": "string",        "start": 0,        "end": 0,        "probability": 0      }    ],    "model": "string",    "media": {      "bytes": 0,      "content_type": "string",      "container": "string",      "width": 0,      "height": 0,      "has_audio": true,      "has_video": true    },    "timing": {      "total_seconds": 0,      "compute_seconds": 0    },    "limits": {      "max_duration_s": 0,      "max_bytes": 0    }  }}