Captions deep-dive

Everything about captions in Pluged AI: generation, import, styling, timing, and troubleshooting.

Captions make your video accessible and engaging. Generate automatically, import from files, style with skins, and fine-tune every word.

Generation

Auto-generate from audio

  1. Click Captions tab in left panel
  2. Select language (or "Auto")
  3. Click Generate captions

Steps:

  • Audio extraction (separate audio from video)
  • Model loading (one-time, caches)
  • Transcription (speech to text with timing)
  • Caption generation (creates text overlay elements)

Time depends on video length (approximately 1-2x runtime).

On-device vs. cloud

| Path | When to use | |------|-------------| | On-device | Local processing, slower but private | | Cloud (Whisper) | Faster, higher accuracy, requires connection |

Toggle in Settings → Captions → Transcription provider.

Importing captions

SRT files

Import existing subtitle files:

  1. Click Captions tab
  2. Click Import → "Import SRT"
  3. Select .srt file
  4. Choose styling options

Learn more: SRT format

ASS files

Advanced SubStation Alpha for styled subtitles:

  1. Click Captions tab
  2. Click Import → "Import ASS"
  3. Select .ass file
  4. Styles may be simplified (not all ASS features supported)

Learn more: ASS format

Caption styling

Each caption element has properties:

Font

  • Family — system fonts, web fonts
  • Size — relative to canvas
  • Weight — normal, bold
  • Style — normal, italic
  • Transform — none, uppercase, lowercase

Colors

  • Fill — text color, solid or gradient
  • Background — box behind text
  • Stroke — outline (weight, color)
  • Shadow — blur, offset, color

Layout

  • Align — horizontal: left, center, right
  • Vertical position — top, center, bottom
  • Safe zone — stays within safe margins for social platforms

Caption skins

Apply complete styles instantly:

| Skin | Use | |------|-----| | tiktok-bold | Bold, high-impact; great for social clips | | minimal-clean | Subtle, elegant; good for education | | boxed-contrast | Solid background box; high accessibility | | editorial-highlight | Underlined highlight bars; premium feel | | neon-pop | Glowing, colorful; gaming/music content | | documentary-lower | Traditional lower-third; interviews |

Apply via agent:

"Restyle captions with tiktok-bold"

"Apply minimal-clean caption skin"

Timing and editing

Edit text

  1. Select caption clip on timeline
  2. Double-click or click Text in properties panel
  3. Edit words
  4. Press Enter to confirm

Adjust timing

  • Drag edges — extend/contract individual caption
  • Drag clip — move entire caption in time
  • Split — divide long captions into multiple

Merge/split

  • Merge — select multiple captions, right-click → "Merge"
  • Split — select caption, position playhead, press S

Word-level timing

Captions generated by Pluged AI have word-level timing:

  • Each word knows its start/end time
  • Animations can highlight per word
  • Syncs to speech precisely

Per-character animation

Some caption skins support animating each character:

  • Stagger — characters appear one by one
  • Speed — chars per second
  • Direction — left-to-right, right-to-left

Troubleshooting

| Issue | Solution | |-------|----------| | Generation fails | Check audio quality; try cloud transcription; verify language | | Timing off | Regenerate or manually adjust clip edges | | Wrong words | Edit text directly; re-transcribe if consistently wrong | | Captions too fast | Split into shorter segments; or merge if too slow | | Styling doesn't apply | Skins require existing captions; generate/import first |

Tips

  • Generate early — captions help with timing agent edits
  • Style per platform — TikTok (bold), courses (minimal), docs (lower-third)
  • Two lines max — easier reading; split long captions
  • Safe zone — keep text within center 80% for social platforms
  • Color contrast — use boxed-contrast over busy backgrounds

See also

PreviousTransitionsNextAudio editing
Community