OpenAI brings Advanced Voice to Mac and Windows desktop apps

Talking to an AI on your phone always felt a bit like making an awkward speakerphone call. But when you are sitting in front of a monitor writing code or digging through research, having a voice in your ear changes the whole workflow.

OpenAI just rolled out its Advanced Voice Mode to ChatGPT desktop apps across macOS and Windows. Previously restricted to Android and iOS, the feature brings low-latency, emotionally responsive voice conversations directly to your computer.

Advanced Voice lands on your workstation

To get started, Mac and Windows users simply need to update their ChatGPT desktop apps to the latest build. A new waveform icon now sits right next to the text prompt box; clicking it kicks off the voice session.

Alongside the desktop rollout, OpenAI introduced five new voices: Vale, Spruce, Arbor, Maple, and Sol. Each comes with distinct pitch, tonality, and regional accents, designed to sound noticeably more human than typical robotic text-to-speech engines.

First demoed during OpenAI's Spring Updates in May and pushed to mobile in September, the model handles voice modulation, mimics human emotions, and allows real-time conversational interruptions.

On the desktop, the use case shifts dramatically toward productivity. Instead of tapping on a phone screen, users can verbally brainstorm paper outlines, debug code out loud, or upload datasets and have a two-way verbal discussion about the findings.

There are clear restrictions, though. The feature remains locked behind ChatGPT Plus and Team paid subscriptions. Furthermore, users based in the EU, the UK, Switzerland, Iceland, Norway, and Liechtenstein are excluded from this rollout.

The race for the desktop heats up

The timing tells a broader story. On the very same day, Anthropic launched its own Mac and Windows desktop apps, laying the groundwork for its Claude Computer Use feature. Meanwhile, Google is reportedly building an agentic AI tool for browsers that can handle real-world errands like purchasing products and booking tickets.

After spending two years battling on web interfaces and mobile apps, major AI labs are realizing that real work still happens on desktop machines.

Why it matters

For power users, developers, and students, having hands-free conversation on your computer is a tangible quality-of-life upgrade. Being able to drop a spreadsheet into ChatGPT and verbally interrogate the data without taking your hands off whatever else you are doing is genuinely useful. If you already pay for Plus or Team, updating today is a no-brainer.

For the broader industry, this confirms the shift toward the desktop as the ultimate battleground for autonomous agents. Mobile operating systems are walled gardens, but desktop environments give AI models access to files, native apps, and deep workflows.

We think OpenAI's execution here is clever for building habit, but it also highlights a limitation. While Anthropic is teaching Claude to actually take over the keyboard and mouse, OpenAI is still keeping ChatGPT confined to a conversational box. Advanced Voice gives ChatGPT a remarkably human throat, but it still lacks desktop hands.