dictation-toolbox / dragonfly

Speech recognition framework allowing powerful Python-based scripting and extension of Dragon NaturallySpeaking (DNS), Windows Speech Recognition (WSR), Kaldi and CMU Pocket Sphinx
GNU Lesser General Public License v3.0
388 stars 75 forks source link

Improve the dictation element recognition decoding procedure #365

Closed drmfinlay closed 2 years ago

drmfinlay commented 2 years ago

This procedure has, until now, only been able to decode recognition results with dictated words differentiated from command words. There are some scenarios where recognition results do not have this characteristic, resulting in recognition failure.

This limitation is done away with here; the procedure now treats all result words properly. The DNS/Natlink behaviour described in issue number #242, and partially fixed in #268, is now properly fixed. The change also removes the need to use uppercase letters to successfully match dictation elements with the text-input ("text") and CMU Pocket Sphinx ("sphinx") engine back-ends.