nicedreamzapp

nicedreamzapp / NarrateClaude

Public

Hands-free on-device voice loop for macOS: Apple SFSpeechRecognizer + cloned-voice TTS, continuous listening, zero cloud. Originally built to drive claude-code-local.

20
2
100% credibility
Found Apr 15, 2026 at 12 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Shell
AI Summary

NarrateClaude is a macOS tool for continuous hands-free voice dictation into a terminal window using Apple's built-in speech recognition, paired with local text-to-speech for offline conversational interactions.

How It Works

1
🔍 Discover NarrateClaude

You find this cool tool on GitHub that lets your Mac listen to you speak and talk back in a voice like yours, all without needing the internet.

2
💻 Set up on your Mac

Download the files to a folder on your Apple Silicon Mac and install the free Apple building tools if needed.

3
🔊 Choose a speaking voice

Create a simple speaking command using your Mac's built-in voice or connect your own cloned voice for replies.

4
🚀 Launch the listener

Run the easy starter script, which opens a window and gets everything ready to hear you.

5
🎙️ Speak and listen

Approve microphone access once, then talk naturally – your words show up, your Mac responds out loud, and it pauses smartly to avoid echoes.

✨ Enjoy hands-free help

Now chat with your Mac while moving around, cooking, or relaxing, keeping everything private and working anywhere.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 12 to 20 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is NarrateClaude?

NarrateClaude turns your Apple Silicon Mac into a hands-free device for continuous voice conversations, using Apple's SFSpeechRecognizer for on-device listening and cloned-voice TTS for replies—all zero cloud, fully offline. You speak naturally into a mic, it types your words into a Terminal window (like for claude-code-local), processes via local AI, and narrates back in your voice. Built in Shell with Swift for the listener, it enables keyboard-free loops for coding or any CLI tool.

Why is it gaining traction?

Unlike cloud voice AIs that leak data and die without Wi-Fi, this runs entirely local with smart endpoint detection that handles noise and auto-pauses during TTS to kill feedback. Developers hook on the privacy, zero costs, and ambient feel—talk while walking, cooking, or stretching, with your Mac narrating steps in real-time. Pairs seamlessly with local models for hands-free github workflows.

Who should use this?

macOS devs running local LLMs like Gemma in claude-code-local, wanting voice-driven code reviews or debugging without typing. Remote workers needing hands-free device setups for multitasking during calls. Ambient computing tinkerers building offline stacks with browser-agent.

Verdict

Try it if you're on Apple Silicon and crave voice-first dev—setup takes minutes via bash scripts, docs are thorough. Low 10 stars and 1.0% credibility score mean it's early and unproven; test permissions and stability first before relying daily.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.