ykdojo

Super Voice Assistant

Development community intermediate

SOURCE https://github.com/ykdojo/super-voice-assistant

Description

macOS voice assistant with global hotkeys - transcribe speech to text with offline models (WhisperKit or Parakeet) or cloud-based Gemini API, capture and transcribe screen recordings with visual context, and read selected text aloud with Gemini Live. Fast, accurate, and simple.

Installation

This entry records only its repository, not the path inside it, so there is no exact command to give. Open the source below and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

README

Super Voice Assistant

macOS voice assistant with global hotkeys - transcribe speech to text with offline models (WhisperKit or Parakeet) or cloud-based Gemini API, capture and transcribe screen recordings with visual context, and read selected text aloud with Gemini Live. Fast, accurate, and simple.

Demo

**Parakeet transcription (fast and accurate):**

https://github.com/user-attachments/assets/163e6484-a3b1-49ef-b5e1-d9887d1f65d0

**Instant text-to-speech:**

https://github.com/user-attachments/assets/c961f0c6-f3b3-49d9-9b42-7a7d93ee6bc8

**Visual disambiguation for names:**

https://github.com/user-attachments/assets/0b7f481f-4fec-4811-87ef-13737e0efac4

Features

**Voice-to-Text Transcription**

Press Command+Option+Z for local offline transcription (WhisperKit or Parakeet)
Press Command+Option+X for cloud transcription with Gemini API
Choose your engine in Settings: WhisperKit models or Parakeet (faster, more accurate)
Automatic text pasting at cursor position
Transcription history with Command+Option+A

**Streaming Text-to-Speech**

Press Command+Option+S to read selected text aloud using Gemini Live API
Press Command+Option+S again while reading to cancel the operation
Sequential streaming for smooth, natural speech with minimal latency
Smart sentence splitting for optimal speech flow

**Screen Recording & Video Transcription**

Press Command+Option+C to start/stop screen recording
Automatic video transcription using Gemini 2.5 Flash API with visual context
Better accuracy for programming terms, code, technical jargon, and ambiguous words
Transcribed text automatically pastes at cursor position

Requirements

macOS 14.0 or later
Xcode 15+ or Xcode Command Line Tools (for Swift 5.9+)
Gemini API key (for text-to-speech and video transcription)
ffmpeg (for screen recording functionality)

System Permissions Setup

This app requires specific system permissions to function properly:

1. Microphone Access

...