# Build a YouTube transcript corpus for RAG

**Use case:** 

Collect transcript text and LLM-ready context from public YouTube videos to support retrieval, indexing, or grounded analysis.

## Input

```json
{
  "videoUrls": [
    "https://www.youtube.com/watch?v=dQw4w9WgXcQ"
  ],
  "channelUrl": "https://www.youtube.com/@mkbhd",
  "searchQuery": "customer support best practices",
  "maxResults": 5,
  "language": "en",
  "includeMetadata": false,
  "outputFormat": "all",
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": [
      "RESIDENTIAL"
    ]
  }
}
```

## Output

```json
{
  "title": {
    "label": "Video title",
    "format": "string"
  },
  "transcriptText": {
    "label": "Full transcript (plain text)",
    "format": "string"
  },
  "wordCount": {
    "label": "Word count",
    "format": "integer"
  },
  "tokenEstimate": {
    "label": "Estimated token count",
    "format": "integer"
  },
  "languageUsed": {
    "label": "Language used",
    "format": "string"
  },
  "isAutoGenerated": {
    "label": "Auto-generated captions",
    "format": "boolean"
  }
}
```

## About this Actor

This example demonstrates how to use [YouTube Transcript and Subtitle Data Scraper](https://apify.com/khadinakbar/youtube-transcript-extractor) with a specific input configuration. Visit the [Actor detail page](https://apify.com/khadinakbar/youtube-transcript-extractor) to learn more, explore other use cases, and run it yourself.