Local transcription
Choose Fast, Balanced, or Accurate speech recognition. After a model’s first download, later transcription runs locally.
Private · Local · No subscription · Windows
Create automatic video captions from speech, correct every word, design the subtitle style, fine-tune timing visually, and export an SRT or burn captions into an MP4—all in one focused Windows app.
One-time purchase · Current version 1.2.0
Polley CaptionsVideo caption editor demo
Watch Polley Captions turn speech into editable text, style captions over playable video, adjust subtitle timing, and export finished results. The workflow runs locally, so you can add text to video without uploading your media to an online caption service.
Watch the Polley Captions demo on YouTubeA complete caption workflow
Choose Fast, Balanced, or Accurate speech recognition. After a model’s first download, later transcription runs locally.
Edit individual words, insert missing words, search and replace, and undo changes without losing the original transcription.
Show one word at a time, compact phrases, or sentences with controls for words, lines, characters, gaps, and duration.
Preview captions over the actual video with playback, seeking, font, color, outline, background, capitalization, and wrapping controls.
Set exact X/Y positions, use social safe-area guides, and layer fades with slide, dissolve, particle, scale, blur, and other in/out transitions.
Save a project, reopen it, revise it, and render again. Export UTF-8 SRT subtitles or an H.264/AAC captioned MP4.
Visual timing editor
Caption blocks sit directly below playback. Drag a whole block to move it, pull either edge to shorten or extend it, zoom in for precision, snap to a grid, and undo or reset timing edits. Automatic smoothing also prevents isolated words from flashing too quickly.
Drop in video or audio.
Choose an accuracy mode.
Polish words and grouping.
Preview style and timing.
Save SRT or captioned MP4.
Supported media
Video: MP4, MOV, MKV, AVI, WebM · Audio: MP3, WAV
No. Media processing and transcription happen locally. Internet is used to download a selected speech model the first time and to check the small official update manifest.
Yes. Save the project file, reopen it later, change corrections, grouping, styling, placement, transitions, or timing, and export again.
The installer may show a Windows reputation warning because it is not code-signed. Always download it from Eric Polley’s official itch.io store page.
Polley Captions can check the official ericpolley.com version file and tell you when a newer release is available. It opens the official product page so you stay in control of downloading and installing it.

A complete local caption workflow for creators who want control without a subscription.
Buy Polley Captions on itch.io