Update README.md

README.md

@@ -188,7 +188,7 @@ The tokenizer for these model was built using the text transcripts of the train
 The model is trained on composite dataset comprising of around 8500 hours of English speech:
 
 - [Librispeech](https://www.openslr.org/12)
-    - Data Collection Method: by Human
+    - Data Collection Method: Automated
     - Labeling Method: by Human
 - [Mozilla Common Voice 11.0 English](https://commonvoice.mozilla.org/en/datasets)
     - Data Collection Method: by Human
@@ -197,10 +197,10 @@ The model is trained on composite dataset comprising of around 8500 hours of Eng
     - Data Collection Method: by Human
     - Labeling Method: by Human
 - [Fisher](https://catalog.ldc.upenn.edu/LDC2004S13)
-    - Data Collection Method: by Human
+    - Data Collection Method: Automated
     - Labeling Method: by Human
 - [MLS](https://www.openslr.org/94/)
-    - Data Collection Method: by Human
+    - Data Collection Method: Automated
     - Labeling Method: by Human
 - [Voxpopuli](https://github.com/facebookresearch/voxpopuli)
     - Data Collection Method: by Human