Skip to content

Releases: brogrammerMW/herdr-voice

herdr-voice 1.4.1

Choose a tag to compare

@brogrammerMW brogrammerMW released this 26 Sep 21:02
bc44125

The orb stays put.

herdr plugin install brogrammerMW/herdr-voice   # reinstall to update
  • The orb no longer disappears. It could vanish while the Herdr window was still on screen (typically with the voice pane hidden behind your zoomed pane) and stay gone. It now checks itself once a second and comes back if anything has hidden it, and it sits at a higher window level: above every app window, the Dock and the menu bar, including zoomed or full-screen terminals.

Full changes: v1.4.0...v1.4.1

herdr-voice 1.4.0

Choose a tag to compare

@brogrammerMW brogrammerMW released this 26 Sep 19:54
6374121

A fully local voice: Local OpenLive mode runs speech recognition, the model and the voice on your Mac, with no audio leaving it and no per-minute fees.

herdr plugin install brogrammerMW/herdr-voice   # new install, or reinstall to update
herdr-voice                                     # start talking (Grok by default)

What's new

  • Local OpenLive mode. Whisper hears you and Kokoro speaks, both on the GPU inside a sandboxed helper that listens only on your Mac; Qwen 3.5 through Ollama drives the same 30 tools, confirmations and reports as the cloud models. It's the first entry in the orb's model menu. It's an advanced feature with a one-time setup from a source checkout (scripts/local-openlive-setup, needs Node 22+, pnpm 10+ and Ollama); until then the menu shows it as "(not set up)". Full guide: docs/local-openlive.md.
  • Warm-up you can see. The orb is amber with the thinking bubble while the local models load, and blue once it's listening.
  • Survives crashes. If the local speech helper dies, the voice relaunches it and reconnects in about 5 seconds.
  • Setup survives updates. Local mode's models and inventory live in ~/.local/state/herdr-voice, outside the plugin folder that reinstalls replace. Existing caches are migrated once, automatically.
  • Quieter turn-taking in local mode. Coughs, clicks and echo tails no longer become turns; sentences that trail off wait for you to finish.
  • Honest refusals. When a confirmation is refused, the voice now says the action was not done, and keeps saying so if asked later.

Known limits

  • Local mode replies about two seconds after you stop talking; Grok is faster. Choose it for privacy, offline use and cost.
  • macOS with WebGPU only. No vision.

Full changes: v1.3.0...v1.4.0

herdr-voice 1.3.0

Choose a tag to compare

@brogrammerMW brogrammerMW released this 26 Sep 00:26
0e79bcc

The voice now stays out of your way: its pane starts hidden behind the pane you're working in.

herdr plugin install brogrammerMW/herdr-voice   # new install, or reinstall to update
herdr-voice                                     # start talking

What's new

  • Starts hidden. When the voice starts, however you open it (herdr-voice, a keybinding, or Herdr's plugin command), the pane you're in is zoomed to fill the tab and the voice runs behind it. The orb shows what it's doing, and it still introduces itself.
  • Show or hide it from the orb. Right-click the orb → Show voice pane to see the transcript, or Hide voice pane to tuck it away again.
  • Prefer it visible? Set HERDR_VOICE_START_HIDDEN=0 in the plugin's config.env (herdr plugin config-dir brogrammermw.herdr-voice).

Full changes: v1.2.0...v1.3.0

herdr-voice 1.2.0

Choose a tag to compare

@brogrammerMW brogrammerMW released this 26 Sep 00:16
c4bbce0

The voice introduces itself when it starts, can split panes right or down, and controls the rest of your Herdr panes by voice.

herdr plugin install brogrammerMW/herdr-voice   # new install, or reinstall to update
herdr-voice                                     # start talking

What's new

  • A greeting when it starts. When herdr-voice starts, or you switch models from the orb, the voice introduces itself in its own words and says it's ready to help, then waits. Reconnects stay quiet.
  • Split right or down. "Start Codex to the right of claude-2" or "open Claude below the dev server" splits that pane and starts the agent there. "Split the build pane down" opens a plain shell pane.
  • Pane and agent control by voice.
    • Zoom, resize, swap, move (to a new tab, another tab, or next to a pane) and rename panes.
    • Rename agents.
    • "What's running in the build pane?" (program names only).
    • "Why isn't that agent showing up?"
    • Closing a pane always asks for your spoken yes, and never closes herdr-voice's own pane.
    • With HERDR_VOICE_SHELL=1, it can run a command in a pane you can see, after your yes.
  • README: clicking the orb mutes or unmutes; right-clicking switches models or quits.

Full changes: v1.1.1...v1.2.0

herdr-voice 1.1.1

Choose a tag to compare

@brogrammerMW brogrammerMW released this 25 Sep 18:48
00a3bf6

Fixes the herdr-voice command after a plugin install.

In 1.1.0 the command linked to the temporary folder Herdr builds plugins in, which Herdr moves afterwards, so herdr-voice wasn't found. It's now a small launcher script in ~/.local/bin that finds the installed plugin wherever Herdr put it.

To update:

herdr plugin install brogrammerMW/herdr-voice

Then type herdr-voice in any Herdr pane. See 1.1.0 for what the command does.

herdr-voice 1.1.0

Choose a tag to compare

@brogrammerMW brogrammerMW released this 25 Sep 18:47
6d9301d

Start the voice by typing herdr-voice.

herdr plugin install brogrammerMW/herdr-voice   # new install, or reinstall to update
herdr-voice setup                               # once: store your API key in the Keychain
herdr-voice                                     # start talking

What's new

  • herdr-voice starts the voice. In a Herdr pane it opens the voice in a pane below and gives your shell back. It tells you if it's already running. --here runs it in the current pane; --provider gemini picks the model.
  • herdr-voice stop ends it; herdr-voice help lists the commands.
  • On your PATH after install. The plugin's build now runs herdr-voice install-command, which links it into ~/.local/bin (Herdr's install preview shows this step). herdr-voice uninstall-command removes the link.
  • README and launch videos. A launch video is in the README, with 16:9 and 9:16 cuts (original house track) in assets/, plus the demo on YouTube.

If herdr-voice isn't found, add ~/.local/bin to your PATH.

Full changes: v1.0.0...v1.1.0

herdr-voice 1.0.0

Choose a tag to compare

@brogrammerMW brogrammerMW released this 25 Sep 15:20
6798763

First release as a Herdr plugin.

herdr plugin install brogrammerMW/herdr-voice

Then herdr plugin action invoke setup --plugin brogrammermw.herdr-voice to store your API key, and herdr plugin action invoke start --plugin brogrammermw.herdr-voice to start talking. macOS 14+, Herdr 0.9.0+, Swift 5.10+ (Xcode or the Command Line Tools).

What it does: realtime voice (Grok, OpenAI or Gemini Live) that talks with you and drives the Claude Code and Codex agents in your Herdr panes. It sends them work, reads their output, answers approvals after your spoken yes, and speaks up when they finish.

  • Create, list, rename and close workspaces, tabs and worktrees by voice.
  • Start an agent in a new worktree in one request.
  • Read or watch any pane ("tell me when the build prints done").
  • Floating orb pinned to the Herdr window; switch models from its right-click menu.
  • Only your speech is streamed. Replies stay short. Risky actions need your spoken yes.

MIT licensed.