Question 1

Qu'est-ce que GPT-Realtime-2 ?

Accepted Answer

GPT-Realtime-2 is OpenAI's new voice model with GPT-5-class reasoning, designed for live conversations where the AI needs to handle complex requests, call tools, and recover from interruptions. It supports a 128K context window and five levels of adjustable reasoning effort, from minimal to xhigh.

Question 2

Combien de langues GPT-Realtime-Translate prend-il en charge ?

Accepted Answer

GPT-Realtime-Translate accepts speech in more than 70 input languages and can translate into 13 output languages in real time. It's designed to keep pace with a speaker without waiting for pauses or completed sentences.

Question 3

Comment sont facturés les nouveaux modèles vocaux ?

Accepted Answer

GPT-Realtime-Translate and GPT-Realtime-Whisper are billed by the minute, while GPT-Realtime-2 is billed by token consumption. All three are available through OpenAI's Realtime API and accessible via WebRTC, WebSocket, and SIP.

Priceline

OpenAI lance des modèles vocaux capables de raisonner, traduire et transcrire en direct

Un résumé percutant.
Chaque jour.

Priceline

OpenAI lance des modèles vocaux capables de raisonner, traduire et transcrire en direct

Un résumé percutant.Chaque jour.

Un résumé percutant.
Chaque jour.