PraxySante/Qwen3-0.6B-ASR-PostTrain-Medical-FR
PraxySante/Qwen3-0.6B-ASR-PostTrain-Medical-FR is a 0.8 billion parameter language model, fine-tuned from Qwen3-0.6B, specifically designed to correct Automatic Speech Recognition (ASR) transcription errors in medical French. It was post-trained on a comprehensive dataset of approximately 16 million medical and oral French sentences, focusing on improving accuracy for drug names, dosages, anatomy, and abbreviations. This model excels in specialized medical transcription refinement, offering enhanced precision for French clinical documentation.
Loading preview...
PraxySante/Qwen3-0.6B-ASR-PostTrain-Medical-FR Overview
This model is a specialized 0.8 billion parameter language model, fine-tuned from the Qwen3-0.6B architecture. Its primary objective is to enhance the accuracy of Automatic Speech Recognition (ASR) outputs within the medical French domain.
Key Capabilities
- Medical ASR Error Correction: Specifically engineered to identify and correct transcription errors in spoken medical French.
- Domain-Specific Accuracy: Improves the recognition of critical medical terminology, including drug names, dosages, anatomical terms, and common abbreviations.
- Targeted Training: Post-trained on a substantial dataset of approximately 16 million medical and oral French sentences, sourced from diverse medical corpora like BDPM, PARCOMED, CAS, MORFITT, QUAERO, Wikipedia, and FineWeb-2.
Good For
- Medical Transcription Refinement: Ideal for applications requiring high accuracy in transcribing medical dictations or conversations in French.
- Clinical Documentation: Useful for improving the reliability of ASR systems used in generating clinical notes, reports, and patient records.
- Specialized French Language Processing: Suited for tasks where precise understanding and correction of medical French are paramount, distinguishing it from general-purpose language models.