A Speech-enabled Fixed-phrase Translator for Healthcare Accessibility

Abstract

In this overview article we describe an application designed to enable communication between health practitioners and patients who do not share a common language, in situations where professional interpreters are not available. Built on the principle of a fixed phrase translator, the application implements different natural language processing (NLP) technologies, such as speech recognition, neural machine translation and text-to-speech to improve usability. Its design allows easy portability to new domains and integration of different types of output for multiple target audiences. Even though BabelDr is far from solving the problem of miscommunication between patients and doctors, it is a clear example of NLP in a real world application designed to help minority groups to communicate in a medical context. It also gives some insights into the relevant criteria for the development of such an application.

Investigating the Medical Coverage of a Translation System into Pictographs for Patients with an Intellectual Disability

“…As regards the medical language, Glyph (Bui et al, 2012) -which is not an AAC applicationautomatically translates patient instructions from text into pictures with NLP and computer graphics techniques. The BabelDr system (Bouillon et al, 2021) proposes pictographs and allows to translate spoken medical utterances in various languages to communicate with migrants and deaf patients in hospitals. For people with ID, de Knegt et al (2016a,b) designed a tool with pictographs, called STOP-ID, to aid the self-reporting of pain (affect, location, intensity and quality).…”

“…In November 2021, there were 1,126 medical pictographs grouped into 45 (sub)categories, such as medical procedures, covid-19, symptoms, etc. For the medical data to translate, we used French sentences of the BabelDr translation system (Bouillon et al, 2021), designed to facilitate communication between physicians and allophone patients. These data include physician questions, patient instructions, and greetings.…”

“…Two experiments were carried out. We first calculated the number of untranslated words on 700 sentence transcripts of real physician questions, recorded with speech recognition of BabelDr system (Bouillon et al, 2021). We also evaluated 700 sentences, called canonicals, linked to each of these transcripts in the BabelDr system.…”

Enhancing Speech Translation in Medical Emergencies with Pictographs

“…These enable doctors to speak freely, with the system linking the recognition result to the closest source-language match that is a clear and explicit variant of the original sentence. This intermediate result can be presented to the doctor for confirmation, and can also be used as the input for translation into the system's target languages (Mutal et al, 2019;Bouillon et al, 2021).…”

“…Literal translation is often problematic as well, since cultural differences may influence the way questions are asked (Halimi et al, 2021). Some recent studies have showed that both patients and doctors tend to prefer a fixed-phrase translator to generic machine translation such as Google Translate (Turner et al, 2019;Panayotou et al, 2019;Bouillon et al, 2017).…”