quriosity-agent/qcut/subtitles

1. Extract audio + upload          ~1min
    ↓
2. Volcengine transcription (with hot words)  ~2min
    ↓
3. Agent auto-proofread            ~3-5min
    ↓
4. Manual review & confirm         Depends on user
    ↓
5. Burn-in subtitles               ~1-2min

# Extract audio
ffmpeg -i "video.mp4" -vn -acodec libmp3lame -y audio.mp3
 
# Upload to uguu.se (temporary file hosting)
curl -s -F "files[]=@audio.mp3" https://uguu.se/upload
# Returns URL like: https://o.uguu.se/xxxxx.mp3

# Dictionary location: <project>/.claude/skills/qcut-toolkit/videocut/subtitles/dictionary.txt
# Script loads it automatically
 
bash ../talk-edit/scripts/volcengine_transcribe.sh "https://o.uguu.se/xxxxx.mp3"

skills
Claude
Agent

const result = JSON.parse(fs.readFileSync('volcengine_result.json'));
const subtitles = result.utterances.map((u, i) => ({
  id: i + 1,
  text: u.text,
  start: u.start_time / 1000,
  end: u.end_time / 1000
}));
fs.writeFileSync('subtitles_with_time.json', JSON.stringify(subtitles, null, 2));

cd subtitles-dir/
node <project>/.claude/skills/qcut-toolkit/videocut/subtitles/scripts/subtitle_server.js 8898 "video.mp4"

ffmpeg -i "video.mp4" \
  -vf "subtitles='video.srt':force_style='FontSize=22,FontName=PingFang SC,Bold=1,PrimaryColour=&H0000deff,OutlineColour=&H00000000,Outline=2,Alignment=2,MarginV=30'" \
  -c:a copy \
  -y "video_subtitled.mp4"

output/YYYY-MM-DD_video-name/subtitles/
├── 1_transcription/
│   ├── audio.mp3
│   └── volcengine_result.json
├── subtitles_with_time.json    # Core file
└── 3_output/
    ├── video.srt
    └── video_subtitled.mp4