OpenAI has unveiled GPT-Live, an innovative voice AI system designed to enhance the fluidity of conversational exchanges by enabling simultaneous speaking and listening capabilities. This new AI operates on a full-duplex architecture, allowing it to interject with natural conversational fillers like “mhmm” or “yeah” while users are still speaking. This capability aims to create a more seamless interaction by minimizing awkward pauses and maintaining the flow of dialogue.
In situations where users present complex queries necessitating internet searches or sophisticated reasoning, GPT-Live can delegate these tasks to a more robust AI model operating in the background while continuing the interaction. Currently, these advanced functions are managed by GPT-5.5, with plans to incorporate cutting-edge models in subsequent updates.
This development comes on the heels of OpenAI’s announcement regarding the upcoming public release of its GPT-5.6 model series, which will occur once it passes an additional cybersecurity review. The GPT-5.6 series features the flagship Sol model, along with Terra and Luna variants, promising to deliver enhanced capabilities upon release.
OpenAI has started to deploy two versions of these new voice models, GPT-Live-1 and GPT-Live-1 mini, to ChatGPT users globally. Furthermore, the company is set to offer GPT-Live through its API, providing developers and businesses the opportunity to create applications that leverage real-time voice AI functionalities.