Which Whisper model should I use?

Use 'tiny' or 'base' for quick drafts on limited hardware, 'small' or 'medium' for a balance of speed and accuracy, and 'large' for the highest quality results if you have 10GB+ of RAM.

Can I transcribe videos in languages other than English?

Yes, the skill supports auto-detection of languages or manual specification using standard codes like 'es' for Spanish or 'fr' for French.

Which transcription formats are supported?

The skill can output transcripts in plain text (TXT), SubRip (SRT), WebVTT (VTT), and detailed JSON with word-level timestamps.

What prerequisites are needed for this skill?

You must have FFmpeg and OpenAI Whisper installed on your system. FFmpeg handles the media processing while Whisper manages the AI transcription.

Video Processor & Transcriber

Name: Video Processor & Transcriber
Author: basher83

bybasher83

•

コンテンツ管理

Processes video files through format conversion, audio extraction, and AI-powered speech-to-text transcription using FFmpeg and Whisper.

概要

This skill empowers Claude to handle complex multimedia tasks directly from the terminal, leveraging FFmpeg for robust video/audio processing and OpenAI's Whisper model for high-accuracy transcriptions. Whether you need to convert legacy formats to web-optimized WebM, extract audio for podcasts, or generate timestamped subtitles in SRT and VTT formats, this utility provides a streamlined workflow with built-in error handling and performance optimizations for various hardware capabilities.

主な機能

Extract audio from video files in multiple formats including WAV, MP3, AAC, and FLAC
Selectable transcription models ranging from 'tiny' for speed to 'large' for maximum accuracy
AI-powered speech-to-text transcription using OpenAI Whisper with multi-language support
5 GitHub stars
Generate timestamped subtitle files in SRT, VTT, or detailed JSON formats
Convert videos to universal MP4 or web-optimized WebM formats using customizable codecs

ユースケース

Converting legacy camera footage into modern, web-ready formats for website embedding
Generating accurate, timestamped subtitles for lecture or interview recordings
Automating the extraction of high-quality audio tracks from video files for standalone distribution

概要

主な機能

Extract audio from video files in multiple formats including WAV, MP3, AAC, and FLAC
Selectable transcription models ranging from 'tiny' for speed to 'large' for maximum accuracy
AI-powered speech-to-text transcription using OpenAI Whisper with multi-language support
5 GitHub stars
Generate timestamped subtitle files in SRT, VTT, or detailed JSON formats
Convert videos to universal MP4 or web-optimized WebM formats using customizable codecs

ユースケース

Converting legacy camera footage into modern, web-ready formats for website embedding
Generating accurate, timestamped subtitles for lecture or interview recordings
Automating the extraction of high-quality audio tracks from video files for standalone distribution