Correct redundancy

Author: denrea

articles/ai-services/speech-service/how-to-custom-speech-display-text-format.md

@@ -103,7 +103,7 @@ A pattern such as `(AB|CD)-(\d)+` would represent constructs like "AB-9" or "CD-
 
 ### Patterns with explicit replacement
 
-The general philosophy is "you show us what the output should look like, and the Speech service figures out how people say it." But this doesn't always work because some scenarios might have quirky unpredictable ways of saying things, or the Speech service background rules might have gaps. For example, there can be colloquial pronunciations for initials and acronyms. For example, `ZPI` might be spoken as `zippy`. In this case, a pattern like `ZPI-\d\d` is unlikely to work if a user says `zippy twenty two`.  For this sort of situation, there's a display text format notation `{spoken>written}`. This particular case could be written out `{zippy>ZPI}-\d\d`.
+The general philosophy is "you show us what the output should look like, and the Speech service figures out how people say it." But this doesn't always work because some scenarios might have quirky unpredictable ways of saying things, or the Speech service background rules might have gaps. For example, there can be colloquial pronunciations for initials and acronyms - `ZPI` might be spoken as `zippy`. In this case, a pattern like `ZPI-\d\d` is unlikely to work if a user says `zippy twenty two`.  For this sort of situation, there's a display text format notation `{spoken>written}`. This particular case could be written out `{zippy>ZPI}-\d\d`.
 
 This can be useful for handling things that the Speech mapping rules but don't yet support. For example you might write a pattern `\d0-\d0` expecting the system to understand that "-" can mean a range, and should be pronounced `to`, as in `twenty to thirty`. But perhaps it doesn't. So you can write a more explicit pattern like `\d0{to>-}\d0` and tell it how you expect the dash to be read.