Update README.md

README.md

@@ -19,7 +19,7 @@ metrics:
 
 # SI-SNR Estimator introduced for the REAL-M dataset
 
-This repository provides all the necessary tools to perform blind SI-SNR estimation on the REAL-M dataset, from mixture signal + estimated sources. While this SI-SNR estimator is introduced with the REAL-M dataset, it can also be used on any speech separation dataset. 
+This repository provides the Separator models to be able to train the blind SI-SNR estimator. While this SI-SNR estimator is introduced with the REAL-M dataset, it can also be used on any speech separation dataset. 
 
 The SI-SNR estimator is trained with a recordings from the WHAMR! and LibriMix datasets. We used a mixture of 9 separators to create the training data on the fly.