[DOCS] Adds missing word to sentence in deploy NLP models docs. (#2675)

(cherry picked from commit 0e910ae)

docs/en/stack/ml/nlp/ml-nlp-deploy-models.asciidoc

@@ -208,7 +208,7 @@ increases the speed of {infer} requests. The value of this setting must not
 exceed the number of available allocated processors per node.
 
 You can view the allocation status in {kib} or by using the
-{ref}/get-trained-models-stats.html[get trained model stats API]. If you to
+{ref}/get-trained-models-stats.html[get trained model stats API]. If you want to
 change the number of allocations, you can use the
 {ref}/update-trained-model-deployment.html[update trained model stats API] after
 the allocation status is `started`.