voice-transcription

Transcribe speech locally with whisper.cpp without cloud services.

3|1|Updated Nov 17, 2025

Search Tags:#automation #voice #transcription #privacy #on-device #whisper.cpp #local-processing

One-click install

npx skills add https://github.com/aldervall/Voicetype --skill voice-transcription

Or copy as Structured Prompt for Agent▼

Please help me install this Agent Skill.
Skill: voice-transcription
Source: https://github.com/aldervall/Voicetype/tree/main/skills/voice
Command: npx skills add https://github.com/aldervall/Voicetype --skill voice-transcription

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, and includes scripts (resource) components.

What problem does it solve?

This Skill enables private, on-device voice transcription using whisper.cpp, so your speech is never sent to the cloud and your data remains under your control.

Core Features & Use Cases

Local transcription: Transcribe speech on your machine using whisper.cpp, with no API keys or cloud services.
Privacy-first: All processing happens locally, ensuring privacy and compliance.
Auto-install & startup: Checks installation, starts whisper server automatically if needed, and downloads models on first use.

Quick Start

Run: python skills/voice/scripts/transcribe.py --duration 5
If VoiceType isn't fully installed, you will be prompted to install via /voicetype-install or run: bash install.sh
Optional: specify duration with --duration 1-30