OpenAI has started rolling out Advanced Voice Mode (AVM) to users in the Plus and Teams tiers of the ChatGPT app, with Enterprise and Education customers set to receive access next week. The much-anticipated audio feature makes interactions with ChatGPT more conversational and natural. In a subtle nod to the delay in releasing these features, the AI can now say “Sorry, I’m late” in over 50 languages.
New Capabilities for Natural Conversations
The Advanced Voice features aim to improve how users interact with ChatGPT by allowing more dynamic conversations. The AI can now handle interruptions, recognize emotional tones, and switch languages more easily.
Initially showcased in May, Advanced Voice Mode was praised for its potential to make interactions feel more human-like. OpenAI has since improved the AI’s ability to understand accents and has worked on reducing glitches that were seen in earlier tests. In addition, users can now take advantage of Custom Instructions, which allow them to guide how ChatGPT responds, and Memory, which lets the AI remember preferences across sessions.
Updated Design and Five New Voices
The design for Advanced Voice Mode has been updated from the animated black dots seen in May to a blue animated sphere. This redesign accompanies the introduction of five new voices: Arbor, Maple, Sol, Spruce, and Vale. These voices join the existing ones—Breeze, Juniper, Cove, and Ember—bringing the total to nine. The names are nature-inspired, likely reflecting the goal of making AI interactions feel more natural.
One voice notably missing is Sky, which was removed after legal challenges from Scarlett Johansson, who claimed it resembled her voice in the movie Her. OpenAI swiftly removed the voice following the legal action.
Missing Features: Video and Screen Sharing
Despite the new voice capabilities, OpenAI has not yet included the video and screen-sharing features shown during the May demo. These would allow GPT-4o to process visual and audio information simultaneously, enabling users to ask questions about things like images or code on their screens. OpenAI has not provided a timeline for when this feature will become available.
Availability and Access
For now, Advanced Voice Mode is being rolled out to Plus and Teams users, while Enterprise and Education customers will gain access next week. However, it is not yet available in regions like the EU, U.K., Switzerland, Iceland, Norway, and Liechtenstein, as OpenAI continues work on compliance in those areas.
The release of Advanced Voice Mode marks another step toward more natural AI interactions. While some features are still missing, OpenAI’s updates to voice capabilities offer users a more engaging and responsive experience. As the technology develops, further improvements, including multimodal capabilities, are expected.
Discover more from Techish Kenya
Subscribe to get the latest posts sent to your email.
