@aaiccee

u2-audio-file-transcriber converts recorded audio to text via unisound asr api, supporting wav, mp3, m4a, flac, ogg formats up to 2 hours and 100mb, with domain optimization for finance and customer service.

med-record-struct extracts and normalizes chinese outpatient follow-up medical records into fine-grained json fields covering history, diagnosis, and treatment plans. includes privacy and data handling commitments.

chronic-disease-review audits outpatient ocr documents for diabetes and hypertension against review rules, producing structured json responses and natural language summaries. requires disease code and ocr array input.

从中文医患对话文本生成门诊初诊病历，输出结构化分段的病历正文（文本）。

重大疾病理赔智能评估（支持 28 种病种）。输入住院病历结构化数据，调用内网评估接口，输出原始 JSON 与自然语言结论（结论 + 证据）。