Learn extra at:
Moreover, the mannequin supplier mentioned that the up to date gpt-realtime mannequin has proven enhancements in following complicated directions, calling instruments with precision, and producing speech that “sounds extra pure and expressive.”
These enhancements, in accordance with Dai, would assist enterprises use the API for enabling low-latency, pure voice interactions for a spectrum of use instances, resembling real-time medical transcription, conversational reserving assistants, customer support for banking, insurance coverage, and telco, and worker enablement throughout main verticals.
Enterprises accessing the mannequin via the API can use two new voices, Cedar and Marin, the mannequin supplier mentioned.