# Iterative LLM-based improvement for French Clinical Interview Transcription and Speaker Diarization

arXiv: https://arxiv.org/abs/2603.00086 · HTML: https://mlsanity.com/paper/2603.00086/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 26 references that are in the map.
- First posted on arXiv 16 Feb 2026.

## Builds on (most cited first, 13 of 13)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [Calibrate Before Use: Improving Few-Shot Performance of Language Models](https://mlsanity.com/paper/2102.09690/) (2021)
- [VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretation](https://mlsanity.com/paper/2101.00390/) (2021)
- [WhisperX: Time-Accurate Speech Transcription of Long-Form Audio](https://mlsanity.com/paper/2303.00747/) (2023)
- [pyannote.audio: neural building blocks for speaker diarization](https://mlsanity.com/paper/1911.01255/) (2019)
- [CHiME-6 Challenge:Tackling Multispeaker Speech Recognition for Unsegmented Recordings](https://mlsanity.com/paper/2004.09249/) (2020)
- [The Sound of Healthcare: Improving Medical Transcription ASR Accuracy with Large Language Models](https://mlsanity.com/paper/2402.07658/) (2024)
- [Lexical Speaker Error Correction: Leveraging Language Models for Speaker Diarization Error Correction](https://mlsanity.com/paper/2306.09313/) (2023)
- [LLM-based speaker diarization correction: A generalizable approach](https://mlsanity.com/paper/2406.04927/) (2024)
- [Word Error Rate Definitions and Algorithms for Long-Form Multi-talker Speech Recognition](https://mlsanity.com/paper/2508.02112/) (2025)
- [AG-LSEC: Audio Grounded Lexical Speaker Error Correction](https://mlsanity.com/paper/2406.17266/) (2024)
- [Acoustic and Machine Learning Methods for Speech-Based Suicide Risk Assessment: A Systematic Review](https://mlsanity.com/paper/2505.18195/) (2025)

## Abstract (arXiv)

Automatic speech recognition for French medical conversations remains challenging, with word error rates often exceeding 30% in spontaneous clinical speech. This study proposes a multi-pass LLM post-processing architecture alternating between Speaker Recognition and Word Recognition passes to improve transcription accuracy and speaker attribution. Ablation studies on two French clinical datasets (suicide prevention telephone counseling and preoperative awake neurosurgery consultations) investigate four design choices: model selection, prompting strategy, pass ordering, and iteration depth. Using Qwen3-Next-80B, Wilcoxon signed-rank tests confirm significant WDER reductions on suicide prevention conversations (p<0.05, n=18), while maintaining stability on awake neurosurgery consultations (n=10), with zero output failures and acceptable computational cost (RTF 0.32), suggesting feasibility for offline clinical deployment, pending validation on larger corpora.