OpenAI WebRTC Audio Session, now with document context
Simon Willison has updated his OpenAI WebRTC audio tool to incorporate document context and the new GPT‑Realtime‑2 model. This model, promoted by OpenAI as having GPT‑5‑class reasoning with a September 30, 2024 knowledge cut-off, is now accessible through Willison's updated application. Users can engage in audio conversations within their browser about provided document content, exploring information in a conversational manner. AI
IMPACT Enables more sophisticated conversational AI interactions with document-aware audio models.