Microsoft Speech Platform - sampling rate and bit depth -
recognition results best if sampling rate , bit depth of audio match training data of system.
so, know exact sampling rate and/or bit depth (and/or stereo/mono) used in microsoft speech platform (newest, if that's important)? , if so, remember got information?
please note using ms speech platform, not sapi. unless both using same training data, that's not same afaik. precise - use this: http://msdn.microsoft.com/en-us/library/microsoft.speech.recognition.speechrecognitionengine.setinputtowavefile%28v=office.14%29.aspx
my first try based upon c++ code example given on page.
the microsoft.speech sr engine doesn't need training (unlike system.speech sr engine), , relatively insensitive sampling rate (will work > 8 khz sampling rate). 16 bit audio preferred, believe work 8 bit audio.
Comments
Post a Comment