Update README.md

README.md

@@ -81,5 +81,5 @@ This is a draft dataset of prosodic minimal pairs.  It includes two stress doubl
 Remaining words are from superb ks.
 *object10* and *present10* tokens lexically have left stress. *object01* and *present01* tokens lexically have right stress.
 
-The data has separate validation splits for each of *object10*, *present10*, *object01* and *present01*.  In each of these splits, all the tokens
+There are separate validation splits for each of *object10*, *present10*, *object01* and *present01*.  In each of these splits, all the tokens
 have the same segmental content and lexical prosody.  Test splits are not here yet.