Amazon launches tool to convert doctor-patient conversation into text

Amazon has expanded its existing 2017 transcription tool, which automatically converts speech into text, to include a conversation between a doctor and a patient. The communication is then automatically inserted as text in the medical file.

The Amazon Transcribe Medical tool is in fact an expanded version of the Amazon Transcribe released in 2017, with the addition of the recognition of all kinds of medical terms. This machine learning tool allows transcriptions of medical consultations between doctors and patients, including the automatic addition of punctuation. The aim is to reduce the time for physicians to spend on administration, such as entering medical reports in particular into electronic patient systems. Amazon states that the tool accurately translates speech into text in real time, but the company does not say how accurately that is.

According to Amazon, accurate medical transcripts are very expensive, time-consuming, or disruptive to patients. The company says the Transcribe Medical is trained on the terminology and style of medical jargon. The tool works on the basis of an API and can be used, for example, in the form of a mobile app on a device with a microphone, after which an entire conversation can be recorded. In addition, the tool can also be used on the computer to quickly process a doctor’s dictation. Any transcripts can be forwarded to the Comprehend Medical tool, which extracts the relevant medical information. Transcribe Medical is currently only available in two regions in the US and users must be Amazon Web Services customers.