The name of a user, info mailbox, or distribution list might not follow the pronunciation rules of the primary language for your system. To increase the likelihood that the speech recognition feature recognizes the name, you can spell the name the way you would pronounce the name.
Speech recognition in both Speech Auto Attendant and Speech Enabled Message Addressing uses the following fields:
First name + Last name
Display name
Pronounceable name
There is no search order between these fields. A match is a match. You must use the Pronounceable name field to increase speech recognition. However, you can use the other fields to increase the recognition rate.
For example, if the primary language of your system is English, spell Dan DuBois as Dan Doobwah.
You can also enter an alternative name for the user. For example, William Bell might also be known as Bill Bell. In the User Properties Web page, if you enter William in the First name field, Bell in the Last name field, and Bill Bell in the Pronounceable name field, the speech engine recognizes both William Bell and Bill Bell.
Each site has a default language setting. You must specify the user name using the language associated with the user site. If an Auto Attendant includes other sites that have different default languages, the callers at Auto Attendant will have to say the names in the language specified to look up a user.
For example, Site 1 has the default site language set to US English. Additionally, Site 1 Auto Attendant includes Site 2 where the default site language is set to Canadian French. The caller has to use French to look up a user in Site 2 or use English to look up a user in Site 1.
Hence, when including other sites in the same Auto Attendant, Avaya recommends that you must ensure that all sites have the same default site language to ensure optimal recognition accuracy and efficiency.