ASR Under the Stethoscope: Evaluating Biases in Clinical Speech Recognition across Indian Languages
NeutralArtificial Intelligence
- A systematic audit of Automatic Speech Recognition (ASR) performance in Indian healthcare settings has been conducted, focusing on languages such as Kannada, Hindi, and Indian English. The study compares various ASR models, including Indic Whisper and Google speech to text, and evaluates transcription accuracy across different demographics, revealing significant performance variability and biases based on speaker roles and language use.
- This evaluation is crucial as it highlights the reliability of ASR technologies in clinical environments, which are increasingly used for documenting patient interactions. Understanding the biases and error patterns can inform improvements in ASR systems, ensuring they serve diverse populations effectively.
- The findings underscore ongoing challenges in ASR technology, particularly in multilingual contexts like India, where language diversity can lead to disparities in healthcare communication. This situation reflects broader issues in AI and language processing, where performance gaps often exist between different languages and dialects, emphasizing the need for more inclusive and context-aware solutions.
— via World Pulse Now AI Editorial System
