Update README.md

README.md

@@ -63,7 +63,7 @@ Here's the output: ڭالت ليا هاد السيد هادا ما كاينش ب
 
 ====================================
 
--v3 (fine-tuned on 10 hours of audio + changed hyperparameters + discovered a huge bug when using the letter ا)
+-v3 (fine-tuned on 10 hours of audio + changed hyperparameters + discovered a huge bug when using the letter ا that improves the WER dramatically)
 
 **Wer**: 23.44
 
@@ -71,7 +71,7 @@ Here's the output: ڭالت ليا هاد السيد هادا ما كاينش ب
 
 **Validation Loss**: 33.92
 
-The validation loss is still high also because the validation data contains words that have never been trained before. The solution is to add more data and more hours of training.
+The validation loss goes down as we add more data for training.
 
 Further training to decrease the training Loss makes this model overfit a little bit.