Your voice now fully controls this AI browser

Perplexity Unleashes Voice-Controlled Comet Browser: Talk Your Way Through the Web

In a bold move that could reshape how we interact with the internet, Perplexity has just rolled out a major upgrade to its Comet browser: a fully voice-controlled interface that lets you navigate the web without ever touching your keyboard or mouse. This isn’t just another voice search feature—it’s a complete reimagining of browser control that’s launching today on desktop and arriving soon for iOS users.

The hands-free revolution is here

Starting immediately, desktop users can activate Comet’s new voice mode by pressing Shift + Alt + V (or Shift + Option + V on Mac). The feature transforms your browser into a conversational interface where you can open websites, scroll through pages, follow links, and perform complex navigation tasks—all through natural speech commands.

Perplexity CEO Aravind Srinivas announced the rollout on X, calling it “the first time you can fully control the browser hands free.” The iOS version is set to arrive within days, potentially making Comet the first mobile browser that truly requires no physical interaction.

Built on cutting-edge AI technology

The voice mode runs on OpenAI’s latest real-time model, specifically the gpt-realtime-1.5 architecture designed for low-latency voice interactions. Srinivas publicly thanked OpenAI’s multimodal team for their contribution to the project. Perplexity claims to have improved tool-call stability by over 25%, meaning the system is significantly more reliable at actually executing your commands rather than just understanding them.

The audio quality has also seen substantial improvements, with better pacing and natural-sounding voices that work well for extended listening sessions. This addresses one of the biggest complaints about voice interfaces: the robotic, unnatural delivery that makes long interactions exhausting.

Why this changes everything

Most browsers treat voice as an afterthought—you speak a query, get results, then tap your way through the rest of the process. Comet wants to eliminate that friction entirely. Imagine saying “scroll down, open the third link, summarize this page, and compare it to what’s in my left tab” and having it all happen seamlessly.

This approach aligns perfectly with the growing trend toward ambient computing, where technology fades into the background and interactions become more natural. Perplexity is betting that the future of web browsing isn’t about clicking and typing—it’s about conversation.

The timing is particularly interesting given the increasing sophistication of AI models and the growing frustration with traditional interface paradigms. Voice control has finally reached a point where it can handle complex, multi-step tasks reliably.

Privacy by design

Unlike many tech companies that collect and monetize your browsing data, Comet takes a different approach. The browser processes voice commands locally when possible and doesn’t store click histories in the cloud by default. There’s no ad tracking or profile building based on your browsing habits. This privacy-first stance could be a major differentiator in a market where data collection is the norm.

What’s coming next

For desktop users, the feature is live now. iOS users can expect the update around March 11, based on App Store pre-order listings. The real test will be whether the voice controls feel natural and reliable during actual use, not just in controlled demos.

Perplexity isn’t stopping there. The company is developing Comet Assistant, an AI that learns your preferences and can help with shopping, food ordering, and flight booking based on your patterns. A password manager and cross-device sync are also in development, though Android users are still waiting on these features.

The iOS voice upgrade represents just the beginning of what Perplexity envisions as a completely reimagined browsing experience—one where you talk to your browser the way you’d talk to a knowledgeable assistant.

voice-controlled browser, hands-free web browsing, AI voice assistant, Perplexity Comet, OpenAI integration, real-time voice technology, browser navigation, voice commands, ambient computing, privacy-focused browsing, desktop voice control, iOS browser update, multimodal AI, tool stability, natural language processing, browser innovation, voice interface, hands-free technology, AI browsing assistant, web navigation voice control

“This is the first browser you can control entirely with your voice”
“Talk to your browser like you talk to a friend”
“The future of browsing is hands-free”
“Voice control that actually works”
“Navigate the web without touching a thing”
“AI-powered browsing just got a major upgrade”
“Privacy-first voice browsing arrives”
“The browser that listens and acts”
“Comet’s voice mode: finally, voice control that delivers”
“Talk, don’t type: the new way to browse”
“Voice browsing that understands complex commands”
“The end of the keyboard and mouse?”
“Perplexity’s bold bet on voice-first computing”
“Browser voice control that’s actually useful”
“Voice navigation that feels natural”
“The browser that learns what you want”
“Voice browsing without the privacy trade-offs”
“AI voice assistant built right into your browser”
“The hands-free web is here”
“Voice control that handles real tasks, not just simple queries”,