[AINews] GPT-Realtime-2, -Translate, and -Whisper: new SOTA realtime voice APIs

OpenAI has launched three new voice models in the Realtime API: GPT-Realtime-2, GPT-Realtime-Translate, and GPT-Realtime-Whisper. The new release features significant enhancements, including a 15.2% performance increase in Big Bench Audio. These models aim to improve usability rather than sound quality, introducing new features like the ability to use preliminary phrases before main responses, which helps developers enhance user interaction. Additionally, the models offer smoother retrieval and better understanding of specialized terminology, making AI interactions more responsive and effective.