Hello all!
I am trying to brainstorm ideas to get around a recognition issue. Number: 8
I've been working with VoiceAttack since 2015, and understand recognition issues very well. One thing I've never been able to get around, and something I avoid using for anything important, is single word recognition of the number 8. I have no issues with any other single number, except for the rare recognition of the alpha version (seven,five,etc.).
I first noticed it tinkering with a voice calculator, but now it's causing issues with a bookmark system I'm creating.
"Create a new bookmark"
"What number will this bookmark be?"
"8"
...crickets...
"Bookmark number _ , set. Is this correct?"
"No!"
"Would you like to try again?"
I've tried creating a dummy command of [1..100] just for recognition purposes, but it never catches the word 8 no matter what I do. I have a General American Accent, so I doubt there is issue there (I sound like yankee newsmen).
Any ideas would be very helpful! I've thought of using a second word that gets dropped (like "Number 8") but when I do that, it ALWAYS uses the alpha version of 8 (Recognized: 'number eight').
Say, '{TXT:TTS_DBS_BookmarkNumberRequest}' (and wait until it completes)
Wait for spoken response: '[1..100]'
Say, 'Bookmark [entry' or 'number] {TXT:~avcs_temp_text_number}, set. Is this correct?' (and wait until it completes)