skills.rest
srgaba
srgabaCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

openai-whisper

Transcribe audio files to text locally using the Whisper CLI.

Updated Mar 30, 2026
Search Tags:#cli#offline#subtitles#transcription#whisper#audio#speech-to-text
One-click install
npx skills add https://github.com/srgaba/open-claw --skill openai-whisper-srgaba
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: openai-whisper
Source: https://github.com/srgaba/open-claw/tree/main/project/skills/openai-whisper
Command: npx skills add https://github.com/srgaba/open-claw --skill openai-whisper-srgaba

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Converting speech recordings into text usually requires uploading audio to a cloud API, which raises privacy concerns and recurring costs. This Skill runs OpenAI's Whisper model locally through its CLI, so audio never leaves your machine and no API key is needed. ## Core Features & Use Cases - Local Speech-to-Text: Transcribe audio files such as MP3 and M4A entirely offline using the whisper command. - Translation and Subtitles: Use the translate task and SRT output format to produce translated subtitles from foreign-language audio. - Model Size Control: Choose smaller models for speed or larger models for accuracy, with models cached in ~/.cache/whisper after first download. - Use Case: Record a meeting as an M4A file, then run whisper with the medium model to generate a plain-text transcript saved alongside the recording. ## Quick Start Transcribe the attached audio file meeting.mp3 to a text file using the whisper CLI with the medium model.

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 to text with Whisper CLI?▼

Run whisper followed by the audio file path, for example whisper /path/audio.mp3 --model medium --output_format txt --output_dir . The transcript is written as a text file in the output directory you specify.

How to generate subtitles from audio using Whisper?▼

Pass --output_format srt to the whisper command to produce a subtitle file instead of plain text. You can combine this with --task translate to create translated subtitles from foreign-language audio.

Does OpenAI Whisper require an API key or internet connection?▼

The Whisper CLI runs models locally and does not require an API key. Internet access is only needed once per model, since models download to ~/.cache/whisper on first run and are reused afterward.

Which Whisper model size should I use for transcription?▼

Use smaller models when speed matters and larger models when accuracy matters. On this install the --model flag defaults to turbo, and you can override it with options like medium for a balance of quality and speed.

Why is the first Whisper transcription run slow?▼

The first run downloads the selected model weights to ~/.cache/whisper, which takes time depending on model size and network speed. Subsequent runs load the cached model and start transcribing immediately.

Related Skills

Explore More Content & Communication Skills→
SecKatieSecKatie

parakeet

Transcribe audio files to timestamped text using parakeet-mlx on Apple Silicon.

Community
Intermediate
VolinadVolinad

youtube-transcript

Fetch YouTube video transcripts in plain text, timestamps, SRT, or JSON.

Community
Intermediate
HybridAIOneHybridAIOne
126

speech.transcribe

Transcribe audio or video into structured text, segments, and metadata.

Official
Intermediate
TikTzukiTikTzuki

Google Cloud Text-to-Speech

Convert text to speech and transcribe audio via cloud APIs.

Community
Advanced