AWS HealthScribe FAQs


AWS HealthScribe is a HIPAA-eligible service empowering healthcare software vendors to build clinical applications that automatically generate clinical notes by analyzing patient-clinician conversations. AWS HealthScribe combines speech recognition and generative artificial intelligence (AI) to help reduce the burden of clinical documentation by transcribing patient-clinician conversations and generating easier-to-review streamlined clinical notes.

Healthcare software vendors can integrate AWS HealthScribe in their applications, allowing clinicians and medical scribes to quickly review and edit preliminary clinical notes to finalize their documentation. AWS HealthScribe also extracts structured medical terms from transcripts that can be used to generate workflow suggestions in your application, improving the user experience and reducing the number of steps.

AWS HealthScribe analyzes patient-clinician conversation audio to: 1) transcribe the consultation, 2) identify speaker roles (for instance, clinician, patient), 3) segment the transcript (for instance, small talk, visit flow management, assessment, treatment plan), 4) extract medical terms (for instance, medication name, medical condition name), 5) summarize notes for key sections of the clinical document (for instance, history of present illness, treatment plan), and 6) create evidence mapping (linking every sentence in the AI-generated note with corresponding transcript dialogues). For more information, visit AWS HealthScribe features.

You can use Amazon Simple Storage Service (Amazon S3) to store both input (audio files) and output (transcription and notes text files) from AWS HealthScribe. AWS HealthScribe uses Amazon Bedrock’s large language model (LLM) capabilities as well as Amazon Comprehend Medical and Amazon Transcribe Medical to generate optimized clinical notes by understanding patient-clinician conversation.

AWS HealthScribe uses conversational and generative AI to transcribe patient-clinician conversations and generate clinical notes. The results produced by AWS HealthScribe are probabilistic and might not always be accurate due to various factors, including audio quality, background noise, speaker clarity, the complexity of medical terminology, and context-specific language nuances. AWS HealthScribe is designed to be used in an assistive role for clinicians and medical scribes, rather than as a substitute for their clinical expertise. As such, AWS HealthScribe output should not be employed to fully automate clinical documentation workflows, but rather to provide additional assistance to clinicians or medical scribes in their documentation process. Please ensure that your application provides the workflow for reviewing the clinical notes produced by AWS HealthScribe and establishes expectation of the need for human review before finalizing clinical notes.

Every AI-generated clinical note includes references to the original consultation transcript to make it easier for users to accurately validate clinical notes provided in your application. Providing these references along with the summary notes to clinicians or medical scribes helps foster trust and encourages safe use of AI in clinical settings.

Request access to AWS HealthScribe. Once allow-listed, you will receive a welcome email and instructions for using the service. 

AWS HealthScribe is available for English-US language in batch (asynchronous) mode.

AWS HealthScribe is available in the US East (N. Virginia) Region.

AWS HealthScribe is optimized for general medicine and the orthopedics specialty.

Privacy and security

AWS HealthScribe is a HIPAA-eligible service.

Requests to the AWS HealthScribe APIs and the console are made over a secure (SSL) connection. You can use AWS Identity and Access Management (IAM) to control which IAM users have access to specific AWS HealthScribe actions and resources. Your trust, privacy, and the security of your content are our highest priority. We implement appropriate and sophisticated technical and physical controls, including encryption at rest and in transit, designed to prevent unauthorized access to, or disclosure of, your content, and we ensure that our use complies with our commitments to you. Please see data privacy for more information.

AWS HealthScribe does not use content processed by the service for any reason other than to provide and maintain the service. The content processed by the service is not used to develop or improve the quality of AWS HealthScribe or any other Amazon ML technologies. Users have full control over their data and determine whether they prefer to store their data in local environments or self-managed cloud storage. You always retain full ownership of your content.

Yes, AWS HealthScribe uses ML and is continually being trained to make it better for your customer use cases. The service does not store or use customer data used in the service to train the models.

AWS HealthScribe does not use content processed by the service for any reason other than to provide and maintain the service. AWS HealthScribe only processes the voice recording to generate AI results and does not retain your data after the AI output is provided to you.

The consultation audio is expunged from our systems once the service is delivered and is only accessible to you. To ensure privacy of your data, the data is also encrypted at rest and in transit to prevent unauthorized access to, or disclosure of, your content. The only drawback is if you require technical support. Because we do not retain your original data, we would not have convenient access to that data should you require the support team to investigate potential service issues. Support is still available, but might not be as expedient because we might need to ask for additional information from you.

Get started with AWS HealthScribe

Explore AWS HealthScribe pricing

Use pay-as-you-go pricing with no upfront costs.

Explore AWS HealthScribe features

Find out how AWS HealthScribe works.

Start using AWS HealthScribe

Begin building with AWS HealthScribe in the AWS Management Console.