by OpenAI
GPT-4o Transcribe Diarize is an automatic speech recognition (ASR) model with built-in speaker diarization, meaning it can associate audio segments in a conversation with different speakers.
| Input | $2.5 / 1M tokens |
| Output | $10 / 1M tokens |
| Context window | 16,000 tokens |
| Modalities | text, audio |
| Features | thinking, function_calling, structured_outputs |
Use gpt-4o-transcribe-diarize via the AIHubMix unified API — one interface for every major LLM.