ChatGPT Plus to Receive Advanced Voice Mode for More Natural Conversations

On Tuesday, OpenAI revealed that paid testers will soon get access to the more sophisticated Voice Mode. The functionality, revealed at an event back in May of this year, leverages…

By Ishika Setia

August 1, 2024

3 min read

Comments

On Tuesday, OpenAI revealed that paid testers will soon get access to the more sophisticated Voice Mode. The functionality, revealed at an event back in May of this year, leverages the GPT-4o AI model. It has real-time answers with an authentic voice along with the power to determine a user’s mood.

This feature will be available in the coming weeks for all ChatGPT Plus users, though OpenAI has yet to provide an update on when additional tools shown off during its showcase – including video and screen-sharing abilities- will go live.

ChatGPT Plus with Advanced Voice Mode

Voice Mode would help users have more natural conversations with the chatbot over X and even allow them to slightly interrupt it or hear a range of voices, said OpenAI in an update post on X (formerly Twitter). A short video introduced how you will be capable to access this feature after it gets switched on.

This short video may indeed reveal that an invite notification will show up at the lower part of a specific group ChatGPT Plus locale urging them to attempt another Voice Mode device in its Beta stage. If you tap the notification, you will be taken to a new page that reads “You’re invited to try the advanced Voice Mode” where you can turn on this feature.

The feature is in the process of rolling out to a limited number of Plus users, but there’s no word on how you become eligible. This alpha rollout is powered by OpenAI’s flagship GPT-4o large language model. Regarding the delay, OpenAI clarified that “Since we first demoed advanced Voice Mode, we’ve been enhancing the safety and quality of voice interactions as we prepare to release this cutting-edge technology to a broader audience.”

These GPT-4o voice capabilities have been rigorously tested by more than 100 external red teamers in 45 languages. Ethical hackers who try and break into systems before the technology is publicly released are called red teamers. Currently, users can access four preset voices after the feature is rolled out, but the controversial “Sky” voice, which was previously similar to actress Scarlett Johansson, has not yet been reintroduced to ChatGPT.