Update README.md

README.md

@@ -14,7 +14,7 @@ datasets:
 
 [FastSpeech 2](https://arxiv.org/abs/2006.04558) text-to-speech model from fairseq S^2 ([paper](https://arxiv.org/abs/2109.06912)/[code](https://github.com/pytorch/fairseq/tree/main/examples/speech_synthesis)):
 - English
-- 200 male/female voices (random speaker in the widget)
+- 200 male/female voices (random speaker when using the widget)
 - Trained on [Common Voice v4](https://commonvoice.mozilla.org/en/datasets)
 
 ## Usage