Related Experiment Video
Updated: May 28, 2026
03:14
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
Automatic Speech Recognition in Healthcare in the Post-LLM Era: A Scoping Review
Maram Alabbad1, Waad Alhoshan1
1Computer Science Department, College of Computer and Information Sciences, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh 11432, Saudi Arabia.
Healthcare (Basel, Switzerland)
|May 27, 2026
Summary
Large Language Models (LLMs) enhance healthcare Automatic Speech Recognition (ASR) for intelligent data processing. While LLM-based ASR reduces documentation time, standardization and equity testing are crucial for safe clinical use.
Area of Science:
- Healthcare Technology
- Artificial Intelligence
- Clinical Informatics
Background:
- Automatic Speech Recognition (ASR) is evolving with Large Language Models (LLMs) beyond transcription.
- LLM-based ASR offers intelligent reasoning, summarization, and structuring of clinical data.
- This review maps the current landscape of LLM-based ASR in healthcare.
Purpose of the Study:
- To scope the emerging field of LLM-based ASR in healthcare.
- To examine applications, technical methods, evaluation strategies, and challenges.
- To provide an overview of recent advancements (2022-2025).
Main Methods:
- Systematic scoping review following PRISMA-ScR guidelines.
- Searched multiple databases for peer-reviewed, open-access studies.
- Included studies published between January 2022 and December 2025.
Main Results:
- Nineteen studies were included, focusing on administrative documentation (42.1%), diagnosis, therapy, and communication.
- Whisper and GPT/LLaMA models were prevalent, with LLMs central in 68.4% of studies.
- Reported documentation time reductions (30-90%) but noted inconsistent privacy reporting, limited equity testing, and fragmented LLM evaluation.
Conclusions:
- LLM-based ASR shows promise for reducing clinical documentation burden.
- Significant gaps exist in evaluation standardization, equity considerations, and reproducibility.
- Further development is needed to ensure safe and equitable clinical deployment.