local-transcript

Extract transcripts from local media files using embedded subtitles or Whisper.

Updated Oct 27, 2025
One-click install
npx skills add https://github.com/SomtoUgeh/dotfiles --skill local-transcript
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: local-transcript
Source: https://github.com/SomtoUgeh/dotfiles/tree/main/claude/skills/local-transcript
Command: npx skills add https://github.com/SomtoUgeh/dotfiles --skill local-transcript

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openai>=1.0.0, and includes scripts (resource) components.

What problem does it solve?

This Skill extracts transcripts from local video and audio files using embedded subtitle tracks when available, or OpenAI Whisper API as a fallback, enabling quick transcription without uploading files to the cloud.

Core Features & Use Cases

  • Tries embedded subtitle tracks first for instant transcripts.
  • Falls back to Whisper transcription when no subtitles are present.
  • Supports optional timestamps and preserves transcript integrity for easy reading and search.

Quick Start

Use uv to run the local transcript tool: uv run scripts/get_local_transcript.py "/path/to/media/file" uv run scripts/get_local_transcript.py "/path/to/media/file" --timestamps

Frequently Asked Questions about local-transcript

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

FAQPage Schema
How do I generate transcripts from local media files without uploading them?

To generate transcripts from local media, this Skill extracts text from embedded subtitle tracks first and falls back to the OpenAI Whisper API, enabling offline transcription directly from your personal video and audio collections.

Can I add timestamps to subtitles extracted from local video and audio?

Yes, you can add timestamps to subtitles extracted from local video and audio by passing the --timestamps flag when running the Python script, preserving transcript integrity for easy reading and searchable archives.

Do I need ffmpeg and an OpenAI API key to transcribe local audio files?

Yes, you need ffmpeg and ffprobe to process local media files, and an OpenAI API key is required for Whisper transcription fallback when no embedded subtitle tracks are present in the source video or audio.

What is the best way to transcribe interviews when the video already has subtitles?

The best way to transcribe interviews with existing subtitles is to use this Skill, which prioritizes extracting embedded subtitle tracks for instant transcripts before attempting slower API-based Whisper transcription.

How do I run the local transcript script using uv?

To run the local transcript script using uv, execute 'uv run scripts/get_local_transcript.py' followed by your media file path, and append '--timestamps' if you need time markers in the output text.

Why does Whisper transcription fallback trigger instead of using embedded subtitles?

Whisper transcription fallback triggers when ffprobe detects no embedded subtitle tracks in the local media file, requiring an OpenAI API key to process the audio and generate a searchable text transcript.