1. Use built-in Voice for a live Codex conversation
Open the ChatGPT desktop app on macOS or Windows, select Codex, then select the Voice control. Allow microphone access when prompted and describe the task or question.
OpenAI documents Voice in Codex as a live interface that can use the tools and permissions available to Codex. That makes it the natural choice for back-and-forth coordination.
2. Treat live Voice and dictation as different interfaces
Live Voice is conversational: you can speak, interrupt, ask questions, and coordinate an active task. Dictation is text-first: speech becomes editable text before you decide what to send.
Choose based on the control point you want. A live conversation is fast; a visible brief is easier to inspect, archive, compare, or reuse across agents.
3. Check the current Voice data controls
OpenAI’s Voice documentation describes retention and model-improvement controls for audio, video, and transcripts. These rules can depend on the voice mode, plan, and workspace, so use the current OpenAI help article rather than a third-party summary.
On IraVoice Local, speech recognition runs on your computer and no audio or text leaves it. The one-time model download and the weekly license check-in are disclosed separately and never carry your words.
4. Use IraVoice for a reviewable, correctly spelled prompt
Click into the Codex prompt in a terminal or the app, hold Right Command on a Mac or Right Ctrl on Windows (Ctrl+Space on Linux), describe the change, and release. IraVoice finds the Git project Codex is working in and spells its code names: “use effect” becomes useEffect and “pipeline dot swift” becomes Pipeline.swift.
IraVoice reads only names, never source text, skips secrets and dependency folders, and sends nothing off the computer. It types the prompt and does not press Enter by default, so you review the text before Codex acts. Codex still owns execution and permissions.