Your voice now has complete control over this AI browser
Perplexity has just launched a new voice feature for its Comet browser. Desktop users can start using it immediately, while iOS users will have to wait a few days. This update allows you to navigate completely hands-free—open websites, scroll through pages, or click links simply by speaking.
On desktop, you can activate it by pressing Shift + Alt + V, or Shift + Option + V on Mac. The same functionality will be available soon on iPhones, potentially making Comet the first mobile browser you can use entirely without touching it.
This feature utilizes OpenAI’s new real-time model. CEO Aravind Srinivas shared the news on X and acknowledged OpenAI’s multimodal team. The aim is to enable full browser navigation through voice commands, transcending just voice searches.
Powered by OpenAI’s latest voice technology
This voice capability is built on OpenAI’s gpt-realtime-1.5 model, designed for efficient voice interaction. Srinivas credited OpenAI’s team, stating that Perplexity has improved tool-call stability by over 25%. This means there are fewer errors when you ask the browser to perform tasks. The voice outputs are also enhanced for better clarity and pacing during extended use.
The importance of complete voice control
Typically, voice functionality in browsers is seen as a novelty; you ask a question, get results, and then click. Comet aims for a fully voice-driven experience, allowing you to inquire about anything displayed on your screen. You could say “scroll down, open the third link, summarize this page, compare it to the tab on the left” without needing to touch a keyboard.
This launch aligns well with the shift towards ambient computing. Perplexity believes interacting with the web through voice will enhance user experience. This feature is already active on desktops.
Comet also takes a unique approach to privacy. It processes voice commands locally whenever feasible and doesn’t default to storing click histories in the cloud, thereby avoiding ad tracking profiles based on your browsing habits.
What to expect in the near future
Desktop users can start testing the voice feature today. For iOS users, it is expected to roll out around March 11 based on App Store pre-order information. The key question remains whether these controls will feel intuitive during real use rather than just in demonstrations.
Perplexity is also expanding its offerings. Comet Assistant is being designed to understand your preferences and assist with shopping, food ordering, or flight searching based on your usual activities. A password manager and cross-device synchronization are also in development, with Android users expecting those features soon. For iOS, this voice upgrade is just the beginning.








