[App][Voice] Add a configurable wake word to Voice settings "The Jarvis Feature"
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 125k
- Forks
- 19.4k
- PR merge metrics
- PR metrics pending
Description
What variant of Codex are you using?
ChatGPT/Codex desktop app on Windows. The feature would also be useful on macOS.
What feature would you like to see?
Please add native wake-word activation under Settings → Voice.
Suggested settings:
- Wake-word activation: Off / On
- Custom wake phrase, for example “Jarvis”
- Allow activation while Codex is running in the background
- Choose whether activation starts Voice in the current context or a new conversation
User problem
I want to start live Codex Voice simply by saying a configurable phrase such
as “Jarvis”.
The current workaround requires configuring a global Codex shortcut and
connecting it to Windows Voice Access or another automation utility. This is
complicated, especially on localized Windows installations, because custom
named Voice Access commands are currently limited to English language variants.
This functionality should be available directly in Codex without requiring a
separate accessibility or automation application.
Expected behavior
When Codex is running in the background:
- The user says the configured wake phrase.
- Codex opens and starts live Voice.
- A clear visual and audible indicator confirms that Codex is listening.
- Ending Voice returns Codex to the configured background state.
Wake-word detection should be explicitly enabled by the user and preferably
processed locally. Audio should not be transmitted before the wake phrase is
detected.
Related request
#19971 concerns dictation recording controls. This request concerns starting
a full live Voice conversation using a configurable wake phrase.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by locating the desktop Voice settings and the existing global shortcut or background-activation entry points. Compare how Windows and macOS handle Voice startup, background state, visual and audible indicators, and local audio processing. Done means an explicitly enabled configurable wake phrase starts the requested Voice context without transmitting audio before detection.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- macos, rust
- Domain
- audio-video-rtc, desktop
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100