Sapana Micro Software
Turn any PDF into a narrated broadcast
PDF → narrated video with synchronized neon highlighting
What it delivers
Broadcast-style document video—narration, spatial highlighting, and camera motion—without a manual edit suite.
Word sync
One active neon highlight tracks the spoken word in real time.
News-style camera
Smooth pan-and-zoom close-ups keep the reader’s eye on the text.
Seconds, not minutes
Unique-frame deduplication renders once per cue, not once per frame.
Pipeline at a glance
Conceptual flow from the technical paper—no implementation details on this site.
Text index
PDFKit extraction with Vision OCR fallback for scanned pages.
Timed speech
Word-boundary cues aligned to synthesized narration audio.
Timeline plan
Speech cues become highlight segments on each page.
Render & encode
Neon overlays, optional pan-and-zoom camera, H.264 + AAC mux.
Headline results
Unique frames vs total
0 / 129
Smoke PDF deduplication
Full paper at 1080p
0s
~6.4 min of narration
Automated tests
0
Timing, mapping, camera
Max camera zoom
0×
Word-mode close-up
| Document | Resolution | Audio | Frames | Wall time |
|---|---|---|---|---|
| 1-page smoke PDF | 640×360 | ~4 s | 129 | ~2 s |
| 1-page smoke PDF | 1920×1080 | ~4 s | 129 | ~2 s |
| Academic paper (CSN) | 1920×1080 | ~6.4 min | ~11,458 | ~36 s |
From the abstract
paper-highlighter converts PDF documents into narrated high-definition video with synchronized neon highlighting and optional news-agency-style camera motion. The pipeline indexes text, exports sample-accurate speech cues, maps highlights to page coordinates, and encodes broadcast-ready MP4—all on macOS using system frameworks. Unique-frame deduplication reduces compositing from every timeline frame to one render per visual state, yielding sub-minute conversion for long documents.
Full paper (PDF) →