retime-captions-to-words

Snap SRT caption starts to nearest whisper word timestamps within tolerance.

6|1|Updated May 29, 2026
One-click install
npx skills add https://github.com/gooseworks-ai/gooseworks-ads-skills --skill retime-captions-to-words
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: retime-captions-to-words
Source: https://github.com/gooseworks-ai/gooseworks-ads-skills/tree/main/skills/atoms/captions/retime-captions-to-words
Command: npx skills add https://github.com/gooseworks-ai/gooseworks-ads-skills --skill retime-captions-to-words

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill corrects timing discrepancies between captions and spoken words in video transcripts, ensuring captions land on word boundaries.

Core Features & Use Cases

  • Caption Timing Correction: Snaps SRT caption cue starts to the nearest word from a whisper word-timestamps file within a specified tolerance.
  • End Time Adjustment: Recomputes end times to preserve original cue duration unless overlap occurs.
  • Quality Checks: Ensures no cue starts after its end, no overlap, and whisper words file is well-formed.
  • Use Case: Before a video goes live, use this Skill to ensure captions are accurately timed with the spoken content.

Quick Start

Run the retime-captions-to-words script with the input SRT file and output path, and provide the path to the whisper word-timestamps JSON file.

Frequently Asked Questions about retime-captions-to-words

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I fix SRT caption timing to match spoken word boundaries?

To fix SRT caption timing, this Skill snaps SRT cue starts to the nearest word timestamps from a whisper word-timestamps file within a specified tolerance. It recomputes end times to preserve the original cue duration unless an overlap occurs.

What is the best way to align SRT cue starts with whisper word timestamps for video post-production?

The best way to align SRT cue starts is running a script that matches them to whisper word timestamps within a set tolerance. This corrects timing discrepancies so captions land accurately on word boundaries before a video goes live.

Do I need whisper word timestamps to correct caption timing in an SRT file?

Yes, you need whisper word timestamps and an SRT file as input to correct caption timing. The Skill relies on the word-level timing data from the whisper file to snap SRT cue starts to the correct word boundaries.

What quality checks are applied when snapping SRT cue starts to word timestamps?

When snapping SRT cue starts, quality checks ensure no cue starts after its end time and no cue overlap occurs. The process also validates that the whisper words file is well-formed before applying timing corrections.

Why do my captions overlap after adjusting SRT start times to word boundaries?

Captions overlap after adjusting SRT start times when recomputed end times intersect subsequent cues. The Skill recomputes end times to preserve original duration but applies quality checks to prevent and manage any overlap.

Can I preserve original SRT cue duration while aligning caption starts to word boundaries?

Yes, you can preserve original SRT cue duration while aligning caption starts. The Skill recomputes end times to maintain the original duration, adjusting only if an overlap with another cue occurs.