OpenAI expands ChatGPT voice assistant to paid users

OpenAI is rolling out its highly anticipated voice assistant feature, known as “advanced voice mode,” to all paid ChatGPT users.

First released in May 2024, this new feature is currently available for ChatGPT Plus and Team users, with Enterprise and Edu users to follow next week. The release comes as OpenAI continues to improve its AI-based chatbot.

Although the company introduced the voice feature at a product launch event in May, the company had to delay the official launch of the feature because of safety issues. In June, the company delayed the feature due to possible problems, especially if the request was a voice impersonation or some copyrighted material. 

By July, only a few users of the ChatGPT Plus paid subscription service had used the tool. OpenAI has since put in place measures such as filters to ensure that the generated music or any other copyrighted audio is not produced. 

OpenAI enhances voice assistant with more options and features

In addition to the general availability, OpenAI has also expanded the range of voice options for the assistant up to nine. Some of the names are nature-inspired, such as Arbor, Spruce, and Maple. This variety helps the user to select the kind of voice they prefer using.

However, there are still some limitations. For example, it lacks the computer vision that was demonstrated earlier this year when people were able to get verbal directions on physical movements such as dancing through their smartphone cameras. Some of these features are not yet included in this release.

Among the key priorities that have been taken into consideration during the design of the assistant is the capability of the voice assistant to mimic some people’s speech. In its statement, OpenAI has noted that the system cannot mimic human beings when it comes to deepfake technology and voice synthesis, given the ethical questions surrounding such technology.

As noted in our earlier post, hackers recently gained access to the OpenAI press account and shared a phishing link that promised “OPENAI” tokens to ChatGPT users. This is the fourth instance of cyber attacks linked to the company’s X account and the fifth cyber attack targeting the American AI research firm since the beginning of this year. Other targets include researcher Jason Wei on September 22, Chief Scientist Jakub Pachocki on June 2024, and CTO Mira Murati on June 2023.