Using context to type a single word that is difficult for the recognizer

[Utterly Voice]

Hello,

We are just looking for input here on a potential upcoming command. There are times when you are editing text, and you want to insert a word that is difficult for the recognizer without proper context. For example, if I want to insert the word "byte", this will most often come out as "bite". However, if I say "byte memory", the word is typed as desired. This is because recognizers rely on context when creating a transcript.

We are thinking about a "context" command that will be called something like this:

"context one byte memory" -> "byte"

The first argument is "one", which indicates which word in the remainder of the utterance should be typed.

Would there be any interest in this command? Are there any suggestions to improve on the idea?