A powerful Apify actor for extracting transcripts/captions from YouTube videos. The actor supports multiple languages, batch processing, and provides both structured JSON with timestamps and plain text formats.

Features 🚀

• Extract transcripts from single or multiple YouTube videos
• Automatic language detection with English preference
• Support for various YouTube URL formats:
  • Regular videos (youtube.com/watch?v=...)
  • Short URLs (youtu.be/...)
  • Live streams (youtube.com/live/...)
• Structured output with:
  • Full transcript text
  • Timestamped segments
  • Available language information
  • Processing status and error handling

Input Parameters 📝

Language Options

Supported languages include: English (en), Spanish (es), French (fr), German (de), Italian (it), Portuguese (pt), Russian (ru), Japanese (ja), Korean (ko), Chinese (zh), Polish (pl), Czech (cs), Slovak (sk), Swedish (sv), Dutch (nl), Hindi (hi), Norwegian (no), Arabic (ar)

Output Format 📊

The actor saves results to its dataset with the following structure:

1{
2    "url": "YouTube video URL",
3    "video_id": "Video identifier",
4    "language_used": "Language code of the extracted transcript",
5    "available_languages": ["array", "of", "available", "language", "codes"],
6    "transcript_json": {
7        "text": "Complete transcript text",
8        "chunks": [
9            {
10                "timestamp": [start_time, end_time],
11                "text": "Segment text"
12            }
13        ]
14    },
15    "transcript_text": "Plain text transcript",
16    "status": "success/error",
17    "error": "Error message if any"
18}

Usage Examples 💡

Input Example

1{
2    "youtube_urls": [
3        "https://www.youtube.com/watch?v=v0sNkmiKsBU"
4    ]
5}

Output Example

1{
2    "url": "https://www.youtube.com/watch?v=v0sNkmiKsBU",
3    "video_id": "v0sNkmiKsBU",
4    "language_used": "en",
5    "available_languages": [
6        "en"
7    ],
8    "transcript_json": {
9        "text": "some people think that people who are so good at math are just nerds and they're whizzes and that's all they know how to do and zack has never been into that category at all...",
10        "chunks": [
11            {
12                "timestamp": [2.639, 5.359],
13                "text": "some people think that"
14            },
15            {
16                "timestamp": [3.76, 7.44],
17                "text": "people who are so good at math are just"
18            }
19            // ... more chunks
20        ]
21    },
22    "transcript_text": "some people think that people who are so good at math are just nerds and they're whizzes and that's all they know how to do...",
23    "status": "success"
24}

Note: The output example shows a truncated version of the transcript for brevity. Actual output will contain the complete transcript with all timestamp chunks.

Limitations ⚠️

• Videos must have captions/transcripts enabled
• Some videos might have restricted caption access
• Quality of transcripts depends on the original captions
• Rate limiting might apply for batch processing

Error Handling 🔧

The actor provides clear error messages when:

• Video URL is invalid
• Transcripts are disabled or unavailable
• Preferred language is not available
• Any other processing errors occur

Support 🤝

For bug reports, feature requests, or questions:

1. Create an issue in the Issues tab on this actor
2. Visit us at knowbase.ai
3. Email us at maciek@knowbase.ai

Changelog 📝

0.1 [Latest]

• Initial release
• Support for multiple video URLs
• Multi-language support
• Structured JSON and plain text output