OpenAI has introduced its Advanced Voice mode for ChatGPT on the web, extending the feature that was previously available only on iOS and Android apps. This announcement, made via a post on X (formerly Twitter), signifies a significant expansion of the conversational capabilities of ChatGPT to desktop users.
Enhancing User Interaction
The Advanced Voice mode aims to make interactions with ChatGPT more natural and conversational. The feature is being rolled out this week to subscribers of ChatGPT Plus, Enterprise, Teams, and Edu. Free-tier users will have to wait a few weeks to access the feature as part of a limited monthly preview.
How to Use Advanced Voice Mode on the Web
Activating the voice feature on the web is straightforward. Users simply need to click the Voice icon in the prompt window’s bottom-right corner and grant their browser permission to access their microphone. Once activated, a blue orb at the center of the screen indicates that voice chat is active.
The voice feature offers nine output voices with unique tones and personalities. OpenAI has introduced five new voices—Arbor, Maple, Sol, Spruce, and Vale—in addition to existing options like Breeze, Juniper, Cove, and Ember. These voices are designed to create a lifelike and engaging dialogue experience, each with distinct emotional tones and the ability to emphasize certain words for more natural interaction.
For instance:
- Arbor is described as “easygoing and versatile.”
- Ember conveys a “confident and optimistic” tone.
This nature-inspired naming reflects OpenAI’s focus on making AI interactions feel smoother and more relatable.
Usage Limits for Subscribers
While paying users gain early access to this feature, there are daily usage limits in place. Plus and Teams subscribers will receive notifications when they have 15 minutes of voice usage left for the day. Free users will receive limited monthly access to test the feature.
Kevin Weil, OpenAI’s Chief Product Officer, noted that these measures are necessary to manage resource availability as the feature scales up.
A Controversial Missing Voice
One notable absence is the previously available “Sky” voice, which was removed following legal and ethical controversies. Critics alleged that Sky bore a striking resemblance to Hollywood actress Scarlett Johansson’s voice, leading to backlash and a lawsuit against OpenAI. Johansson’s legal representatives claimed that the company lacked permission to use a voice that closely mimicked her own. In response, OpenAI suspended this feature in May 2024.
OpenAI stated that any resemblance was unintentional, but internal comments referencing the film Her, where Johansson voiced an AI assistant, added fuel to the controversy. The company has since refrained from including voices that could lead to similar disputes.
What’s Next for Advanced Voice Mode
With this introduction of Advanced Voice mode on the web, OpenAI is taking another step toward making AI interactions feel more personal and accessible. The feature promises to transform user experiences by combining natural dialogue capabilities with versatile voice options.
As the rollout continues, OpenAI is expected to refine this feature further, eventually extending it to all ChatGPT users while addressing concerns around voice replication and ethical usage.
Future Enhancements
OpenAI plans ongoing improvements based on user feedback and technological advancements. The goal is not only to enhance user experience but also to ensure compliance with legal standards regarding voice replication.
In summary, with Advanced Voice mode now available on web platforms, OpenAI reinforces its commitment to creating engaging AI experiences that cater to diverse user preferences while navigating complex ethical landscapes associated with voice technology.