SpeakySpeak
start some Claudes, walk away hear replies as they arrive
A rainbow means replies are waiting.
What the demo replies say
- Terminal 1, events-site: “Deploy finished, the events site is live and the newsletter went out to 212 readers.”
- Terminal 2, markemark: “I found the memory leak. It was the image cache holding every thumbnail forever. One line fix, tests are green, ready for your review.”
- Terminal 3, everething: “The refactor is done, all 34 tests pass and the branch is pushed. The tricky part was the session cache. My notes are in the commit body.”
Free, for Claude Code on your Mac.
The voice runs there too, offline.
See it work
This is how I work: three projects, each with a builder Claude and a side Claude. Press Start and look away. The clock runs fast.
It's like iTunes for your Claude sessions. Each session is a playlist, and they all share one queue.
Turn your sound on.
events-site builder + side
> ship today's newsletter build
> /side watch the deploy
markemark builder + side
> find the preview pane leak
> /side plan the cache cap
everething builder + side
> refactor the search index
> /side check the plan
Example windows. The side Claude only reads and plans. Its replies are silent in this demo; the app reads them too.
Two views of the same app: same queue, same reply, same moment.
Hidden. It pops back up with the next reply.
Drag it next to any terminal. Click the menu-bar icon for the full deck.
Waiting for replies…
events-site
Usage
Status
Checked 10:40 AM, every 5 minutes
Tap ⇅ to turn grouping off. The bottom strip shows Claude usage, status and version (example numbers).
Two Macs, two voices
Running Claude on two Macs with one pair of AirPods? The Macs take turns, so their replies never talk over each other. Give each one its own voice and you know which Mac is talking without looking.
Mac one
voice: Lily
speakingwaits its turn
one pair of AirPods
Mac two
voice: George
speakingwaits its turn
They find each other over Tailscale. Pick each Mac's voice in Settings ▸ Voice. Setup is in the guide.
Claude asks out loud
When a session stops to ask "may I run this?", a chime plays and you hear which one is waiting.
The voice runs on your Mac
No cloud, no API keys, no subscription. Free and MIT licensed.
Your Claude builds it
Paste one prompt and your own Claude builds it from source. After that it updates itself and keeps any changes you made.
What it does
One voice at a time
Replies queue up instead of talking over each other.
AirPods stem controls
Click to pause, skip ahead, or step back.
It waits for you
Paused? New replies line up until you're back.
The mini player
Shows up while a reply plays.
What's new
What I've added lately, newest first. The rest is in the changelog, and the app tells you when there's an update.
- 1.2.27
A pause stays a pause
Paused when the app updated? It comes back paused at the same spot, and new replies wait.
- 1.2.16
Hear which session is talking
Turn on "Say the session name first" and each reply opens with its name.
- 1.2.8
Permission prompts, out loud
When Claude stops to ask "may I run this?", a chime plays and you hear which session is waiting.
Why a voice?
Because eyes are usually busy and ears usually aren't.
ADHD-friendly
You hear each reply when it's done. No staring at a terminal, no forgotten tabs.
Blind and low-vision devs
Every reply read aloud in a natural voice, without steering a screen reader through terminal output.
Hands full
Making dinner or pacing with a coffee? You still hear every reply.
RSI and limited mobility
AirPods clicks and one hotkey handle pause, skip and back. Less mouse, less window-hunting.
A suggested way to work
None of this is required. It's how I work.
A window per project, each its own color.
Name your sessions with /rename.
Clear the list before you walk away.
Start long runs, then walk away.
Two Macs? A different voice for each.
Tab colors are an iTerm2 setting: Session › Edit Session › Tab Color. More in How to use it and the full guide.
Every feature
The queue
- One at a time: replies never talk over each other
- Newest plays next: or group by session and hear each one's replies in the order they happened
- Pause holds everything: new replies wait; a rainbow glow reminds you
- Skip one reply: drop it and go to the next
- Survives restarts: the queue comes back when the app relaunches
- Clear played / clear all from the full deck
- Play now, remove, or move up: one reply or a whole session
Playback
- Speed selector: 0.8× to 2×, right on the full deck
- Volume slider, and every reply plays at the same loudness
- Back 10 / forward 10, restart and previous reply
- Mute that keeps queueing: replies wait until you're back
- Skip hotkey (⌃⌥⌘Space): skips the reply that's playing, from any app, no permission prompts
The voice
- Kokoro voice: a free neural voice that runs on your Mac (Apple Silicon), with dozens to pick from. English only for now.
- Fast: a reply becomes audio in under a second
- Automatic fallback to the best installed macOS voice (Intel works too)
- Voice preview in Settings; a new voice also applies to replies already queued
- Per-machine voices: tell your Macs apart by ear
AirPods & media keys
- Stem clicks: click to pause, double for next, triple for back
- Now Playing: Control Center scrubber while speaking, released when idle so your music keys come back
- Two Macs, one pair: the two apps take turns instead of fighting over your AirPods
What it reads
- The whole reply: every part of a long turn, start to finish
- Tuned for the ear: emoji and markdown stripped, paths spoken as filenames, URLs as "link"
- You choose when: at the end of a reply, or as a long run works, so you hear a request made twenty minutes in
- Permission prompts, out loud: a chime, then which session is asking and what it wants to run
The app
- Floating mini player: auto-appears while speaking, drag it anywhere, in three looks: Classic, Glassy or Glassier
- Accent color: one slider recolors the whole app, white, gray and black included
- Menu-bar icon: orange while it speaks, a line through it when muted, an orange dot when replies are waiting
- Works where Claude Code runs on your Mac: terminal, the desktop app, VS Code, Cursor. Not Claude Code on the web, and not sessions on another machine over SSH
- Native Settings window: it writes plain text files in your home folder, so edits from the shell stay in sync
- Built-in updates: an ↑ on the menu-bar icon when a new one is out, one button in Settings to install it
- Fully local: no accounts, no telemetry, MIT-licensed, works offline
Install
There's no download. Your Claude builds the app on your Mac, checking each step, and can change anything you like later. Just ask it.
Needs macOS 15.6 or newer, Claude Code, Homebrew and the Xcode command-line tools.
How to use it · GUIDE.md · INSTALL.md · github.com/radam5000/speakyspeak
Want a different voice or speed? Add it to the prompt.
Also by Adam
Previews Markdown right in Finder, with the spacebar.
An offline knowledge tablet.