parakeet

Transcribe audio files offline using NVIDIA Parakeet TDT models.

41|3|Updated Jan 2, 2026
One-click install
npx skills add https://github.com/tdimino/claude-code-minoan --skill parakeet-tdimino
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: parakeet
Source: https://github.com/tdimino/claude-code-minoan/tree/main/skills/integration-automation/parakeet
Command: npx skills add https://github.com/tdimino/claude-code-minoan --skill parakeet-tdimino

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a completely offline, local speech-to-text solution, eliminating the need for cloud APIs and ensuring data privacy for transcribing audio files or dictating voice input.

Core Features & Use Cases

  • Local Transcription: Utilizes NVIDIA Parakeet TDT for high-accuracy speech-to-text processing on your own machine.
  • Multiple Modes: Supports transcribing audio files, live dictation directly into text fields (via Handy app), and terminal dictation.
  • Use Case: Dictate meeting notes directly into your editor, transcribe a long audio interview without uploading it, or convert a batch of audio files to text for analysis.

Quick Start

Use the parakeet skill to transcribe the audio file located at '/path/to/meeting.mp3'.

Frequently Asked Questions about parakeet

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

FAQPage Schema
How do I transcribe an audio file completely offline?

Local speech-to-text transcription is handled by NVIDIA Parakeet TDT models running directly on your machine. This ensures complete data privacy by processing audio files entirely offline without needing cloud APIs.

Can I use live dictation directly into my text editor?

Yes, live dictation is supported via push-to-talk applications like the Handy app. You can dictate meeting notes directly into your editor or use terminal-based voice input across 25 languages with automatic detection.

Does offline speech-to-text work with multiple languages?

Offline speech-to-text supports 25 languages with automatic detection. The local models process audio on your machine to identify and transcribe spoken language automatically without manual configuration.

What is the best way to dictate voice input without an internet connection?

The best way to dictate offline is using local speech-to-text models like NVIDIA Parakeet TDT. This approach eliminates cloud API dependencies, ensuring your voice data remains completely private on your own machine.

Do I need cloud APIs to transcribe long audio interviews?

No, you do not need cloud APIs to transcribe long audio interviews. You can use local transcription engines like NeMo CLI or transcribe-rs to process audio files entirely offline on your own machine.