ChatGPT's Voice Mode lets you talk to it out loud and hear responses back, instead of typing. Here's exactly where every option lives and how to use it, on both mobile and desktop.
Enabling Voice Mode
This part shows exactly where the voice button sits and what happens when you tap it.
Voice conversations now happen directly inside your ongoing chat — no separate voice-only screen needed. You can speak, watch responses appear in real time, and still see earlier messages, images, or maps. On the web, click the Voice icon at the bottom-right of the prompt window, then grant microphone permission when your browser asks.
This picture shows exactly where the microphone icon sits inside the message box — bottom-right corner, next to the attach (paperclip) icon.
Example: Step-by-Step Setup
This example lists the full setup flow, matching the labeled picture above.
- Open a chat and look at the message input box at the bottom.
- Tap the microphone icon on the right side of the box.
- Allow microphone access when your browser or app asks for permission.
- You'll see a blue orb appear — this means it's listening and ready.
- Choose from nine output voices, each with a different tone and character, such as "Arbor" (easygoing) or "Ember" (confident).
Note: If you'd rather use the older, separate voice-only screen instead of the blended chat view, this can be switched back in Settings.
Using Voice on Mobile vs Desktop
This part shows exactly where the option lives on each platform, since the icon location differs slightly.
On mobile, go to the app, tap the headphone icon, and enable Voice — you can also turn on the camera to add Vision, so ChatGPT responds to what it sees in real time. On desktop, the feature works the same way inside the browser or app — click the Voice icon next to the input bar, which shows a pulsing blue orb once it's ready.
Comparison Table: Mobile vs Desktop Voice Mode
This table shows exactly where to find the voice option on each platform, side by side.
| Platform | Where to Enable | Extra Option |
| Mobile app | Headphone icon in the chat bar | Camera + Vision for real-time responses |
| Desktop/Web | Microphone icon in the prompt box | Nine voice options to choose from |
Example: On mobile, a user cooking hands-free can tap the headphone icon and ask "How much salt does this recipe need?" without touching the screen again.
Limitations of Voice Mode
This part covers the practical restrictions you should know about before relying on voice for everything.
Voice Mode isn't unlimited, and it has a few real constraints depending on your plan.
- Daily time limits: Plus and Team users are subject to a daily limit for Advanced Voice Mode, and OpenAI notifies you when you have 15 minutes left for the day.
- Free tier restrictions: Free users only get a monthly preview of the feature rather than full daily access.
- Requires microphone permission: Without granting mic access in your browser or phone settings, voice mode won't activate at all.
- Internet-dependent: Voice mode needs an active internet connection — it doesn't work offline like some phone assistants.
Comparison Table: Voice Mode Limits by Plan (H3)
This table summarizes what each plan actually gets access to.
| Plan | Access Level |
| Free | Monthly preview only |
| Plus/Team | Daily limit, with a 15-minute warning near the end |
Conclusion
Voice Mode now lives directly inside your regular chat — tap the microphone (web) or headphone icon (mobile) in the message bar to start talking, and ChatGPT responds both out loud and on screen. Desktop and mobile both support it, with mobile adding an optional camera-based Vision mode for real-time visual responses. Free users get limited monthly access, while Plus and Team plans include daily voice time with a warning near the cutoff.