OUTCOMESAI/MEDIQA-CHAT-2023-Benchmark

MEDIQA-CHAT-2023基准测试包含一个数据集，用于分析医疗提供者（如医生）与患者之间的对话，以及临床笔记生成任务。数据集主要关注反映真实世界医疗场景的多轮对话及其对应的结构化临床笔记。数据集包含两个主要字段：`dialogue`（医生与患者的多轮对话转录）和`clinical_note`（从对话中提取或生成的结构化医疗笔记）。该数据集支持多种应用，包括对话理解、临床笔记生成、语音合成、自动语音识别评估等。

The MEDIQA-CHAT-2023 Benchmark contains a dataset structured for analyzing conversations between healthcare providers (e.g., doctors) and patients, as well as for clinical note generation tasks. It focuses on multi-turn dialogues that mirror real-world medical scenarios and corresponding structured clinical notes. The dataset includes two primary fields: `dialogue` (multi-turn transcriptions of conversations between doctors and patients) and `clinical_note` (structured medical notes extracted or generated from the dialogue). It supports a range of applications in medical NLP, speech processing, and AI evaluation, including dialogue understanding, clinical note generation, voice synthesis, and automatic speech recognition evaluation.