desktop-transcriber-patterns

Community

Patterns for fast Windows desktop transcription.

Authorfreekmetsch
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Desktop transcription on Windows often suffers from latency, accuracy issues, and brittle workflows when integrating audio capture, VAD, model loading, and UI automation. This Skill provides expert patterns to build a robust, real-time transcription app using faster-whisper, with guidance on CUDA setup, VAD tuning, multilingual support, system tray controls, and clipboard paste automation.

Core Features & Use Cases

  • Real-time transcription with CUDA-accelerated models (large-v3, turbo)
  • VAD tuning for push-to-talk and noisy environments
  • Multilingual support with code-switching and initial_prompt vocabulary
  • System tray integration and global hotkeys
  • Clipboard paste automation with Pyperclip/PyAutoGUI

Quick Start

Start the desktop-transcriber-patterns workflow by loading the large-v3 model, configuring 16 kHz real-time audio capture, and activating the tray icon to begin transcription via the hotkey.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: desktop-transcriber-patterns
Download link: https://github.com/freekmetsch/transcriber/archive/main.zip#desktop-transcriber-patterns

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.