Social Voice Video On Demand Transcription

Social Voice Video on Demand Transcription is a connector that transcribes TikTok, Instagram, and YouTube videos from URLs, adding speech-to-text transcripts and video metadata directly to your documents.

This component transcribes videos from TikTok, Instagram, and YouTube using Social Voice. It reads the video URL from each incoming document, sends it for processing, and adds the transcription result back to the same document under social_voice. You can use Datastreamer’s shared token or provide your own Social Voice bearer token.

Component Configuration

Users must specify the Target URL field, which is the JSON path to the property in the incoming document that contains the video URL to process (for example, content.url).

When the Social Voice API returns the result, the connector attaches it to the original document in a social_voice property.

Example

The final document structure looks like this:

{
  "content": {
    "url": "https://www.youtube.com/watch?v=ys47U5aCVDY"
  },
  "social_voice": {
    "video_id": "6a3168c0930d3da007411d98",
    "process_id": "e54add76-d844-465d-b5e4-365bbc0ed9a1",
    "meta_data": {
      "title": "My Top 3 Everyday Sunscreens",
      "platform": "youtube",
      "url": "https://www.youtube.com/watch?v=ys47U5aCVDY"
    },
    "transcription": {
      "status": "ok",
      "lang": "en",
      "baseTranscription": {
        "text": "Well guys, it's here in a blink of an eye...",
        "segments": [
          {
            "id": 0,
            "start": 0.0,
            "end": 7.44,
            "text": "Well guys, it's here in a blink of an eye..."
          }
        ]
      },
      "englishTranscription": {
        "text": ""
      }
    }
  }
}