Names and Speech Recognition

The speech recognition system uses the values you enter in name fields to identify entities (users, groups, sites, locations, and address book entries).

Any name that a user speaks in a badge command must appear in one of the name fields so the speech recognition system can identify it.

Each of these entities has one or more primary name fields (for example, First Name and Last Name are the primary name fields for a user; Group Name is the primary name for a group) as well as one or more Alternate Spoken Name fields.

Use the primary name fields to enter the most common name of an entity. This name should be the one that most people use, not the "official" name. For example, if everyone calls the user "William Bones" by the name "Billy", you should enter "Billy" in the First Name field and "Bones" in the Last Name field.

If people often refer to users by something other than their primary names, such as a nickname or a title, you must also provide an alternate spoken name. For example, if people often call the user "Jason Crunch" by the nickname "Captain", you must enter "Captain Crunch" as his first alternate spoken name.

Similarly, if the user Rob Shostak is a doctor, enter his first and last names as usual, then enter "Doctor Shostak" as one alternate spoken name, and "Doctor Rob Shostak" as another alternate spoken name. This practice accommodates peers who call him "Rob Shostak", as well as nurses who typically call him "Doctor Shostak" or "Doctor Rob Shostak".

Because you cannot enter any letters except the 26 English letters in the Administration Console or User Console, you may sometimes have problems with user names. For example, you cannot enter common non-English characters such as ñ and é. If a user's name is "Louis Céline", you need to enter an approximation such as "Celine" in the console. In addition, if he uses the French pronunciation, or if other users refer to him with this pronunciation, you may need to add an alternate spoken name such as "Saline".

In general, however, do not provide alternate spoken names to accommodate a name that appears to have a "non-English" pronunciation, unless testing indicates that the system cannot recognize it. Vocera understands a wide variety of name pronunciations, and unnecessary alternate names increase the size of the database. An exception to this rule is slang or jargon that is often pronounced differently than it is spelled.