talking-head-recut

Synchronize HTML graphic overlays with video using word-level transcript timestamps.

Updated May 6, 2026
One-click install
npx skills add https://github.com/johan-Rm/agents --skill talking-head-recut-johan-rm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: talking-head-recut
Source: https://github.com/johan-Rm/agents/tree/main/skills/talking-head-recut
Command: npx skills add https://github.com/johan-Rm/agents --skill talking-head-recut-johan-rm

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires ffmpeg, ffprobe, hyperframes, and includes references (resource) and assets (resource) components.

What problem does it solve?

This Skill solves the problem of static, unengaging talking-head videos by automating the synchronization of designed graphic overlays—such as lower-thirds, data callouts, and kinetic titles—with the spoken transcript.

Core Features & Use Cases

  • Automated Transcription: Uses local Whisper to generate word-level timestamps for precise sync.
  • Dynamic Graphic Packaging: Renders HTML-based graphic cards onto video using GSAP animations.
  • Use Case: Transform a 10-minute raw interview into a high-retention social media clip by automatically overlaying key takeaways, data visualizations, and speaker titles synced to the audio.

Quick Start

Use the talking-head-recut skill to package the video file located at path/to/video.mp4 with designed graphic overlays.

Frequently Asked Questions about talking-head-recut

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

FAQPage Schema
How do I automate graphic overlays for talking-head videos using transcript timestamps?

To automate graphic overlays for talking-head videos, this Skill synchronizes HTML-based designs with word-level transcript timestamps. It generates precise sync data using local Whisper transcription and renders GSAP-driven animations via the hyperframes CLI onto your local video content.

Can I use local Whisper transcription to sync kinetic titles with podcast video?

Yes, you can use local Whisper transcription to sync kinetic titles with podcast video. The process generates word-level timestamps from your audio, allowing the system to automatically align dynamic graphic cards and speaker titles directly with the spoken content.

Do I need ffmpeg and hyperframes installed to render HTML graphics onto video?

Yes, you need ffmpeg, ffprobe, and the hyperframes CLI installed to render HTML graphics onto video. ffmpeg processes the local video files, while hyperframes drives the GSAP rendering required to overlay designed graphic cards onto the media.

What is the best way to package interview videos for social media with dynamic data callouts?

The best way to package interview videos for social media is using automated graphic synchronization. This approach overlays designed lower-thirds, data callouts, and kinetic titles onto the video by aligning them with word-level transcript timestamps to create high-retention clips.

Does this automated video editing approach support different aspect ratios for social media?

Yes, this automated video editing approach supports diverse aspect ratios for social media. It is designed to handle interview, podcast, and talking-head video packaging across various formats by rendering adaptable HTML-based graphic overlays.