Benchmarking von ASR-Modellen im deutschen medizinischen Kontext: Eine Leistungsanalyse anhand von Anamnesegesprächen

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Schuster, Thomas, Trögele, Julius, Döring, Nico, Krüger, Robin, Hoffmann, Matthieu, Friedrich, Holger
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908793078546432
author Schuster, Thomas
Trögele, Julius
Döring, Nico
Krüger, Robin
Hoffmann, Matthieu
Friedrich, Holger
author_facet Schuster, Thomas
Trögele, Julius
Döring, Nico
Krüger, Robin
Hoffmann, Matthieu
Friedrich, Holger
contents Automatic Speech Recognition (ASR) offers significant potential to reduce the workload of medical personnel, for example, through the automation of documentation tasks. While numerous benchmarks exist for the English language, specific evaluations for the German-speaking medical context are still lacking, particularly regarding the inclusion of dialects. In this article, we present a curated dataset of simulated doctor-patient conversations and evaluate a total of 29 different ASR models. The test field encompasses both open-weights models from the Whisper, Voxtral, and Wav2Vec2 families as well as commercial state-of-the-art APIs (AssemblyAI, Deepgram). For evaluation, we utilize three different metrics (WER, CER, BLEU) and provide an outlook on qualitative semantic analysis. The results demonstrate significant performance differences between the models: while the best systems already achieve very good Word Error Rates (WER) of partly below 3%, the error rates of other models, especially concerning medical terminology or dialect-influenced variations, are considerably higher.
format Preprint
id arxiv_https___arxiv_org_abs_2601_19945
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Benchmarking von ASR-Modellen im deutschen medizinischen Kontext: Eine Leistungsanalyse anhand von Anamnesegesprächen
Schuster, Thomas
Trögele, Julius
Döring, Nico
Krüger, Robin
Hoffmann, Matthieu
Friedrich, Holger
Computation and Language
Artificial Intelligence
Automatic Speech Recognition (ASR) offers significant potential to reduce the workload of medical personnel, for example, through the automation of documentation tasks. While numerous benchmarks exist for the English language, specific evaluations for the German-speaking medical context are still lacking, particularly regarding the inclusion of dialects. In this article, we present a curated dataset of simulated doctor-patient conversations and evaluate a total of 29 different ASR models. The test field encompasses both open-weights models from the Whisper, Voxtral, and Wav2Vec2 families as well as commercial state-of-the-art APIs (AssemblyAI, Deepgram). For evaluation, we utilize three different metrics (WER, CER, BLEU) and provide an outlook on qualitative semantic analysis. The results demonstrate significant performance differences between the models: while the best systems already achieve very good Word Error Rates (WER) of partly below 3%, the error rates of other models, especially concerning medical terminology or dialect-influenced variations, are considerably higher.
title Benchmarking von ASR-Modellen im deutschen medizinischen Kontext: Eine Leistungsanalyse anhand von Anamnesegesprächen
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2601.19945