Voice recognition: an enabling technology for modern health care?

Area of Science:

• Medical Informatics
• Human-Computer Interaction
• Speech Technology

Background:

• Recent advancements in voice recognition (VR) systems have renewed interest in their application for clinician-mediated computing.
• Affordable, large vocabulary, speaker-independent VR systems show promise for improving usability and effectiveness.

Purpose of the Study:

• To identify and analyze the significant cognitive and implementation challenges hindering the ubiquitous adoption of voice recognition in clinical settings.
• To project the future potential of voice recognition in medicine, contingent on overcoming identified barriers.

Main Methods:

• Literature review and analysis of current voice recognition technology capabilities and limitations.
• Identification and categorization of cognitive factors (e.g., individual differences, cognitive load, cognitive style) and implementation factors (e.g., training, portability, privacy, environment suitability).

Main Results:

• Key challenges include imperfect recognition accuracy, discrete input requirements, and throughput limitations.
• Significant cognitive issues involve accommodating user variability and managing cognitive load.
• Implementation hurdles encompass training needs, portability, privacy concerns, and environmental suitability.

Conclusions:

• Overcoming these cognitive and implementation challenges is crucial for voice recognition to become a widespread input modality in healthcare.
• The continued improvement in VR performance, alongside the resolution of these issues, heralds a new era for voice recognition in medical applications.