Update README.md

README.md

@@ -67,7 +67,7 @@ The model is particularly well-suited for applications requiring detailed transc
 
 WER is computed on LangAge and ESLO, containing interjections, hesitations. This dataset is challenging for ASR models, because it contains many recordings of older people with different voice quality, and because features of spontaneous speech are difficult to transcribe.
 
-Lower WER indicates better performance. 
+Lower WER indicates better performance. For a fair comparison between Whisper Large V3 and FrWhisper, punctuation and capitalization were ignored for computing WER.
 
 | Dataset Category | Sample Size | Whisper Large V3 | FrWhisper | Improvement |
 |-----------------|-------------|------------------|----------------|-------------|
@@ -82,7 +82,6 @@ Lower WER indicates better performance.
 ### Key Performance Highlights
 
 - **14.18 percentage points** overall WER improvement compared to Whisper Large V3
-- **Median WER for LangAge data**: Improved from 100% to 44.44%
 - **Consistent performance** across both training and test data (no overfitting)
 
 ## Training Details