88Z AI, 11 September — OpenAI announces the API launch of GPT-Live-1, a voice model supporting full-duplex conversation and complex scenario handling, suitable for telephony agents.
At a glance
- GPT-Live-1 supports full-duplex voice interaction, handling interruptions and noise
- Voice functions are integrated into a single model to reduce latency, with complex reasoning handled by backend text models
The update
OpenAI has officially integrated its real-time voice model, GPT-Live-1, into its API services. This model features full-duplex conversation capabilities, allowing simultaneous voice input and output while effectively handling interruptions, pauses, and background noise. These features make it particularly suitable for agent scenarios requiring telephony interactions, such as restaurant reservations and customer service. Architecturally, GPT-Live-1 integrates speech understanding and generation within a single model to significantly reduce response latency. For tasks involving complex logical reasoning, the system can delegate processing to backend text models to ensure a combination of accuracy and efficiency.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- IT之家Source published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.