mlx-whisper

Transcribe and translate local audio files using MLX Whisper on Apple Silicon.

1|Updated Feb 15, 2026
One-click install
npx skills add https://github.com/rustykuntz/rocie-tools --skill mlx-whisper-rustykuntz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mlx-whisper
Source: https://github.com/rustykuntz/rocie-tools/tree/main/skills/mlx-whisper
Command: npx skills add https://github.com/rustykuntz/rocie-tools --skill mlx-whisper-rustykuntz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires mlx-whisper, and includes references (resource) components.

What problem does it solve?

This Skill provides a fast and private way to convert spoken audio into text directly on your Apple Silicon Mac, without relying on cloud APIs or requiring API keys.

Core Features & Use Cases

  • Local Transcription: Transcribe audio files (MP3, M4A, etc.) into plain text.
  • Subtitle Generation: Create SRT subtitle files for videos.
  • Language Translation: Translate spoken audio from other languages into English.
  • Model Selection: Supports various MLX Whisper models, from tiny to large, allowing a balance between speed and accuracy.
  • Use Case: You need to transcribe a confidential meeting recording without sending the audio to an external service.

Quick Start

Transcribe the audio file 'meeting_notes.m4a' to text using the mlx-whisper skill.

Frequently Asked Questions about mlx-whisper

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

FAQPage Schema
How do I transcribe audio locally on Apple Silicon without an API key?

Local speech-to-text transcription on Apple Silicon uses MLX Whisper to process audio files like MP3 and M4A directly on-device, eliminating the need for external API calls or keys.

Can I generate SRT subtitle files from audio on my Mac?

Yes, you can generate SRT subtitle files from audio on your Mac using local speech-to-text. The MLX Whisper Skill processes spoken audio on Apple Silicon to create text or subtitle outputs.

Does local speech-to-text work for translating audio from other languages to English?

Local speech-to-text does work for translating audio from other languages to English. The MLX Whisper models process spoken audio on-device to translate foreign language recordings into English text.

What is the best way to transcribe confidential meeting recordings privately?

The best way to transcribe confidential meeting recordings privately is using local speech-to-text. Processing audio on-device with MLX Whisper ensures sensitive data never leaves your Apple Silicon Mac.

Do I need an Apple Silicon Mac to use MLX Whisper for transcription?

Yes, you need an Apple Silicon Mac because MLX Whisper is specifically optimized for Apple Silicon architecture to perform efficient on-device speech-to-text processing without external API calls.

How do I balance speed and accuracy when choosing MLX Whisper models?

You balance speed and accuracy by selecting from various MLX Whisper models ranging from tiny to large. Smaller models provide faster local transcription, while larger models offer greater accuracy for speech-to-text processing.