Medication Errors: Will Statistical Pronunciation Models Help?

Dr. Lucian Galescu

Approximately 12.5 percent of the medication errors reported to the FDA are a result of confusion between drug names, which can have a direct and serious health consequence to a patient. These errors are often caused by confusion between drug names with similar pronunciation (phonetic similarity) or similar spelling (orthographic similarity).

In years past, I developed techniques for mapping orthography and pronunciation to be used for training statistical pronunciation models for speech recognition and speech synthesis. In this talk I will present some ideas about how such pronunciation models may help reduce the incidence of drug name confusions. Time and context permitting I will also speculate about other ways speech and dialogue technology could assist healthcare providers.