askrobots / dbbasic-video
Public
Video analysis: download, keyframes, transcripts, thumbnail grids and AI summaries of what's on screen.
get_transcript.py · 2.3 KB · 80 lines
| 1 | #!/usr/bin/env python3 |
| 2 | """ |
| 3 | Download YouTube transcript |
| 4 | """ |
| 5 | |
| 6 | import argparse |
| 7 | import json |
| 8 | from youtube_transcript_api import YouTubeTranscriptApi as yt_api |
| 9 | |
| 10 | |
| 11 | def download_transcript(video_id, output_file=None): |
| 12 | """Download transcript for a YouTube video.""" |
| 13 | print(f"Fetching transcript for video ID: {video_id}") |
| 14 | |
| 15 | try: |
| 16 | # Get transcript |
| 17 | fetched = yt_api().fetch(video_id) |
| 18 | transcript = [{'start': s.start, 'duration': s.duration, 'text': s.text} for s in fetched] |
| 19 | |
| 20 | if not output_file: |
| 21 | output_file = f"{video_id}_transcript.json" |
| 22 | |
| 23 | # Save JSON version |
| 24 | with open(output_file, 'w') as f: |
| 25 | json.dump(transcript, f, indent=2) |
| 26 | |
| 27 | print(f"Transcript saved to {output_file}") |
| 28 | |
| 29 | # Save readable text version |
| 30 | text_file = output_file.replace('.json', '.txt') |
| 31 | with open(text_file, 'w') as f: |
| 32 | for entry in transcript: |
| 33 | timestamp = entry['start'] |
| 34 | text = entry['text'] |
| 35 | f.write(f"[{timestamp:.2f}s] {text}\n") |
| 36 | |
| 37 | print(f"Text version saved to {text_file}") |
| 38 | |
| 39 | # Print summary |
| 40 | total_duration = transcript[-1]['start'] + transcript[-1]['duration'] |
| 41 | print(f"\nTranscript stats:") |
| 42 | print(f" - Entries: {len(transcript)}") |
| 43 | print(f" - Duration: {total_duration:.2f}s ({total_duration/60:.2f} min)") |
| 44 | print(f" - Total text length: {sum(len(e['text']) for e in transcript)} characters") |
| 45 | |
| 46 | return transcript |
| 47 | |
| 48 | except Exception as e: |
| 49 | print(f"Error: {e}") |
| 50 | return None |
| 51 | |
| 52 | |
| 53 | def main(): |
| 54 | parser = argparse.ArgumentParser( |
| 55 | description="Download YouTube video transcript" |
| 56 | ) |
| 57 | parser.add_argument( |
| 58 | "video_id", |
| 59 | help="YouTube video ID (e.g., 'xAji90wPgec') or URL" |
| 60 | ) |
| 61 | parser.add_argument( |
| 62 | "-o", "--output", |
| 63 | help="Output JSON file (default: {video_id}_transcript.json)" |
| 64 | ) |
| 65 | |
| 66 | args = parser.parse_args() |
| 67 | |
| 68 | # Extract video ID from URL if needed |
| 69 | video_id = args.video_id |
| 70 | if 'youtube.com' in video_id or 'youtu.be' in video_id: |
| 71 | if 'v=' in video_id: |
| 72 | video_id = video_id.split('v=')[1].split('&')[0] |
| 73 | elif 'youtu.be/' in video_id: |
| 74 | video_id = video_id.split('youtu.be/')[1].split('?')[0] |
| 75 | |
| 76 | download_transcript(video_id, args.output) |
| 77 | |
| 78 | |
| 79 | if __name__ == "__main__": |
| 80 | main() |