GPT-4o speech-to-text transcription model with support for audio and text inputs. Provides high-quality transcription with optional prompts for context.