up

README.md

@@ -33,7 +33,7 @@ The model was evaluated on both short and long-form transcriptions, using in-dis
 
 Note that Word Error Rate (WER) results shown here are [post-normalization](https://github.com/openai/whisper/blob/main/whisper/normalizers/basic.py), which includes converting text to lowercase and removing symbols and punctuation.
 
-All evaluation results on the public datasets can be found [here]().
+All evaluation results on the public datasets can be found [here](https://drive.google.com/drive/folders/1iJ5GXQap8Bz_Tn_mh58EfCb81UQHvgzi?usp=sharing).
 
 ### Short-Form Transcription
 
@@ -380,7 +380,7 @@ print(result["text"])
 
 ## Training details
 
-We built a French speech recognition dataset of over 22,000 hours of annotated and semi-annotated speech. After decoding this dataset through Whisper Large V3 and filtering out segments with WER above 20%, we retained approximately 10,000 hours of high-quality audio.
+We built a French speech recognition dataset of over 22,000 hours of annotated and semi-annotated speech. After decoding this dataset through Whisper-Large-V3 and filtering out segments with WER above 20%, we retained approximately 10,000 hours of high-quality audio.
 
 | Dataset | Total Duration (h) | Filtered Duration (h) <20% WER |
 |---|---:|---:|