- Release information added
Update from 22 October 2024:
OpenAI's Advanced Voice Mode is now available for all Plus users in the EU, Switzerland, Iceland, Norway, and Liechtenstein. Previously, access for users from these countries was only possible via VPN.
Original article from 26 September 2024
OpenAI has rolled out Advanced Voice Mode for ChatGPT Plus and Team users. However, many previously demonstrated features are still missing, and the feature is not available in the EU.
OpenAI CEO Sam Altman announced that the rollout of Advanced Voice Mode for ChatGPT Plus and Team users is complete. The new function allows voice interactions with the AI assistant and promises improved accents and faster conversation speeds.
New features include Custom Instructions, Memory, and five new voices. However, many capabilities shown during the introduction of GPT-4o, the model behind Advanced Voice Mode, are not yet available.
During the GPT-4o presentation, OpenAI demonstrated much more extensive functions, particularly in visual processing. These included real-time analysis of videos or graphics and emotion recognition in faces. The currently available Advanced Voice Mode cannot process images or videos.
In the demos, OpenAI showed how the mode could teach a child mathematics by analyzing their drawings on an iPad. Singing or humming, which was also demonstrated in the presentations, is not available in the current version.
Currently, Advanced Voice Mode cannot search the internet. It's also not possible to switch from a text conversation to Advanced Voice Mode or transfer a standard conversation to the advanced mode.
Missing features and no access in all countries
OpenAI also presented a range of multimodal capabilities of the new model during the GPT-4o introduction, including image generation. The company has not yet released these functions either.
Additionally, Advanced Voice Mode is not available in the EU, UK, Switzerland, and some other European countries. A possible reason for this could be the system's ability to recognize emotions - a function that would be prohibited under the EU AI Act.
Since the introduction of Advanced Voice Mode in May, other companies have also introduced or released similar products, particularly Google with Gemini Live. With Moshi, there's also a first - albeit weaker - open-source alternative.
