skills.rest
compozy
compozyOFFICIAL
Explore all skills by this author →
Website𝕏@compozyaiGitHub

hyperframes-media

Generate TTS audio, word-timestamp transcripts, and transparent overlays for HyperFrames.

182|8|Updated Apr 4, 2026
Search Tags:#captions#transcription#tts#background removal#hyperframes#composition assets
One-click install
npx skills add https://github.com/compozy/agh --skill hyperframes-media-compozy
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hyperframes-media
Source: https://github.com/compozy/agh/tree/main/.agents/skills/hyperframes/hyperframes-media
Command: npx skills add https://github.com/compozy/agh --skill hyperframes-media-compozy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It turns raw text and media into the ready-to-composite assets HyperFrames compositions need, saving you from manual narration, caption timing, and background cutout workflows.

Core Features & Use Cases

  • Text-to-speech narration (TTS): Generate spoken audio locally with Kokoro, choose from many voices, and cache models on first run.
  • Audio/video transcription with word timestamps: Produce a normalized transcript.json for caption overlays and downstream use.
  • Background removal for transparent overlays: Create cutout foreground (and an optional inverse “plate” layer) for transparent video/photo compositing.
  • Use Case: Generate a voiceover from a script, transcribe it to get word-level timestamps for captions, and remove the background from a presenter clip to place them over custom graphics.

Quick Start

Use hyperframes-media to create narration, transcribe it, and then generate transparent cutout assets by running it as a local-first pipeline.

Frequently Asked Questions about hyperframes-media

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

FAQPage Schema
How do I generate word-level timestamps for video captions locally?▼

Produce word-level timestamps for video captions by running Whisper-based transcription on your audio or video files. This generates a normalized transcript.json containing precise timing data for caption overlays and downstream composition.

Can I do text-to-speech narration offline without sending my script to a cloud API?▼

Generate text-to-speech narration offline using local Kokoro models. The pipeline caches models on first run, allowing you to produce spoken audio from text scripts without relying on external cloud APIs.

What is the best way to remove video backgrounds for transparent overlays?▼

Remove video backgrounds for transparent overlays by processing presenter clips through background removal. This outputs alpha-friendly video formats, creating a cutout foreground layer and an optional inverse plate layer for compositing.

How do I chain text-to-speech audio generation with transcription to create captioned voiceovers?▼

Chain text-to-speech and transcription to create captioned voiceovers by running a local-first pipeline. Generate TTS audio from your script, then feed that audio directly into Whisper transcription to get word-timestamp captions.

Does Whisper transcription accidentally translate non-English audio into English?▼

Whisper can accidentally translate non-English audio into English if language handling is not enforced. This preprocessing pipeline enforces correct language settings to avoid unintended English translation of your original audio.

Related Skills

Explore More Design & Creative Skills→
geezerrrrgeezerrrr
117

openai-whisper

Transcribe audio and video files into text or SRT subtitles locally with Whisper CLI.

Community
Basic
ZardLi1115ZardLi1115
4

hyperframes

Render HTML and GSAP motion graphics into MP4/WebM video via FFmpeg.

Community
Intermediate
cosmicstack-labscosmicstack-labs
364

hyperframes-media

Generate local TTS audio, word-timestamp transcripts, and transparent overlays for media compositions.

Official
Intermediate
wesleysimpliciowesleysimplicio
1

hyperframes-media

Preprocess HyperFrames media assets to generate TTS narration, transcripts, and background-removed overlays.

Community
Intermediate