PraxySante/Qwen3-0.6B-SFT-ASR-Correction-FR-v2
PraxySante/Qwen3-0.6B-SFT-ASR-Correction-FR-v2 is an 0.8 billion parameter language model, fine-tuned from PraxySante/Qwen3-0.6B-ASR-PostTrain-Medical-FR with a 32768 token context length. This model is specifically trained using Supervised Fine-Tuning (SFT) to correct Automatic Speech Recognition (ASR) outputs in French. It is designed for applications requiring improved accuracy in transcribing spoken French, particularly in specialized domains.
Loading preview...
Model Overview
PraxySante/Qwen3-0.6B-SFT-ASR-Correction-FR-v2 is an 0.8 billion parameter language model developed by PraxySante. It is a fine-tuned iteration of the PraxySante/Qwen3-0.6B-ASR-PostTrain-Medical-FR model, specifically optimized for correcting Automatic Speech Recognition (ASR) errors in French.
Key Capabilities
- ASR Correction: Specialized in refining and correcting outputs from ASR systems for French language inputs.
- Supervised Fine-Tuning (SFT): The model has undergone SFT using the TRL framework, indicating a focus on specific task performance.
- French Language Focus: Designed and trained for applications within the French linguistic context.
Training Details
This model was trained using Supervised Fine-Tuning (SFT) with the TRL framework (version 1.5.0). The development utilized Transformers (version 5.9.0), Pytorch (version 2.12.0), Datasets (version 4.8.5), and Tokenizers (version 0.22.2).
Good For
- Improving the accuracy of French ASR transcripts.
- Applications requiring post-processing of spoken French text.
- Use cases where precise French language understanding from speech is critical.