How long can I record audio for?

The default recording duration is 5 seconds, but you can specify custom durations up to 30 seconds using the --duration parameter.

Does this work with all Whisper models?

By default, the skill downloads a standard optimized model for whisper.cpp during the first-time installation to ensure a balance between speed and accuracy.

What are the prerequisites for using this skill?

The skill includes an automatic setup script that handles dependencies like the Python virtual environment and Whisper binary. You only need a working system microphone.

What happens if the Whisper server isn't running?

The transcription script includes an auto-start feature that detects if the whisper.cpp server is offline and launches it automatically before recording.

Is my voice data sent to any external servers?

No, all transcription is performed locally on your machine using whisper.cpp and a local server to ensure complete privacy and data security.

Voice Transcription

Name: Voice Transcription
Author: aldervall

byaldervall

•

生産性とワークフロー

Enables high-speed, local voice recording and transcription directly within the terminal for hands-free coding and communication.

概要

The Voice Transcription skill provides a privacy-focused solution for developers to input information using their voice instead of typing. By utilizing a local whisper.cpp server, it allows for the recording and transcription of complex issue descriptions, audio notes, or dictated code changes without sending sensitive audio data to the cloud. It features automatic installation detection, managed dependencies, and seamless integration with Claude Code to streamline workflows where verbal communication is more efficient than manual entry.

主な機能

2 GitHub stars
Seamless integration with system microphones for terminal-based voice input
Real-time transcription status and integrated error handling
Local speech-to-text processing using whisper.cpp for maximum privacy
Automatic installation and dependency management for the Whisper server
Configurable recording durations for quick notes or longer explanations

ユースケース

Dictating complex bug reports or architectural explanations to Claude instead of typing
Quickly converting recorded audio snippets into actionable text or documentation
Hands-free terminal interaction when manual typing is slow or inconvenient

概要

主な機能

2 GitHub stars
Seamless integration with system microphones for terminal-based voice input
Real-time transcription status and integrated error handling
Local speech-to-text processing using whisper.cpp for maximum privacy
Automatic installation and dependency management for the Whisper server
Configurable recording durations for quick notes or longer explanations

ユースケース

Dictating complex bug reports or architectural explanations to Claude instead of typing
Quickly converting recorded audio snippets into actionable text or documentation
Hands-free terminal interaction when manual typing is slow or inconvenient