jarvis-voice

Overview

Turn your AI into JARVIS with voice, wit, and personality. This skill uses sherpa-onnx offline TTS and ffmpeg audio processing to give your agent a British Alan voice with metallic audio effects.

Key Features

• Voice output using sherpa-onnx offline TTS
• Metallic audio effects via ffmpeg
• Four research-backed comedy patterns for humor
• Custom webchat styling for purple italic text

How It Works

This skill uses a hybrid output approach, where the user hears the voice first and then sees the transcript. The jarvis shell command is used to execute the audio, and the visible transcript is rendered in markdown with a custom CSS class.

Use Cases

• Session greetings and farewells
• Delivering results or summaries
• Responding to direct conversation
• Any time the user’s last message included voice/audio