OpenAI Brings Real-Time Voice Chats to ChatGPT on the Web

OpenAI has launched Advanced Voice Mode on the web, letting users talk to ChatGPT right from their browser. Announced by Chief Product Officer Kevin Weil on X, the feature allows real-time voice chats, making interactions easier and more personal. The new voice mode follows its debut in ChatGPT’s mobile apps for iOS and Android back in September. Now, web users can enjoy real-time conversations with ChatGPT directly from their browsers. To start a voice conversation, users simply need to click the Voice icon in the prompt window and allow microphone access. A blue orb will appear to signal the start of the conversation. What makes this feature unique is its use of GPT-4o’s native audio capabilities, which allow ChatGPT to understand and respond to non-verbal cues, such as your speaking speed. The chatbot can also show emotion in its responses, making interactions more engaging. In addition to speaking naturally, ChatGPT offers a variety of nine different voices for users to choose from. For example, users can pick "Arbor," which is easygoing and versatile, or "Ember," which is confident and optimistic. This adds another layer of personality to the conversation. The feature is rolling out this week to Plus, Enterprise, Teams, and Edu subscribers. However, there are a few limits: paying users have daily voice usage limits, with notifications when nearing the 15-minute mark. Free users will get a monthly preview to try the voice feature but with limited access. OpenAI plans to extend this feature to free users in the coming weeks, bringing more flexibility to how we interact with ChatGPT.  With these updates, ChatGPT is becoming more than just a chatbot—it’s turning into a conversational partner that can respond to you with voice, personality, and real-time understanding.