openai-whisper

Transcribe local audio files to text or subtitles using the Whisper CLI.

386k|81.1k|Updated Nov 24, 2025

Search Tags:#whisper #audio transcription #speech-to-text #local processing

Ready to run

npx skills add https://github.com/steipete/clawdis --skill openai-whisper

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: openai-whisper
Source: https://github.com/steipete/clawdis/tree/main/skills/openai-whisper
Command: npx skills add https://github.com/steipete/clawdis --skill openai-whisper

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires whisper.

What problem does it solve?

This Skill provides offline speech-to-text capabilities, allowing you to transcribe audio files without internet connectivity or external API dependencies.

Quick Start

Transcribe the audio file 'interview.mp3' using the medium model and save the output as text format.

Core Features & Use Cases

Local Transcription: Process audio files on your device without sending data to external servers.

Quick Start

Transcribe 'podcast_episode.m4a' and translate it to English using the translate task.

Frequently Asked Questions about openai-whisper

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

FAQPage Schema

How do I transcribe audio files to text locally without API calls?▼

Local transcription converts speech to text using Whisper CLI on your device. Install Whisper, run it on audio files (mp3, wav, m4a), and save output as text or subtitle formats without sending data to external servers or requiring internet connectivity.

What audio formats does Whisper support for transcription?▼

Whisper transcribes common audio formats including mp3, wav, and m4a files. Output can be generated as plain text, SRT subtitles, or VTT subtitles, with optional translation to English in offline environments.

Can I translate audio to English while transcribing?▼

Yes, Whisper supports translation as an optional task alongside transcription. Use the translate task parameter to convert speech in other languages to English text output while processing audio files offline.

How do I choose which Whisper model to use for transcription?▼

Whisper allows model selection via the --model parameter to balance speed and accuracy. The default turbo model is cached locally to ~/.cache/whisper, and you can specify alternative models based on your performance needs.

Do I need an internet connection to transcribe audio with Whisper?▼

No, Whisper performs offline transcription entirely on your local device. After initial model download and caching, you can transcribe and translate audio files without internet connectivity or external API dependencies.

Can I save transcripts in different formats from the same audio file?▼

Yes, Whisper supports multiple output formats including plain text (txt), SRT subtitles, and VTT subtitles. Specify output format and directory preferences when running transcription to generate transcripts in your preferred format.