Currently, confidence scoring is only supported in
recognizer configurations that use the lex tree
linguist. Confidence support should be such that it
works regardless of the linguist used.
This should have a higher priority. Confidence for command and control is important. If my grammer is (turn on the) (light | fan ) and sphinx matches this when I say, " my book is a cow" that is not very helpful unless we have confidence score.
This was originally posted in 2004, any plans to add it?
Ticket moved from /p/cmusphinx/feature-requests/18/
Log in to post a comment.
Sign up for the SourceForge newsletter:
You seem to have CSS turned off.
Please don't fill out this field.