Remember to teach them about the "Calibrate" UI in the speech
preferences. Also let the kiddies know that if the recognizer does
not recognize them, then they should not repeat the command louder,
should not overarticulate, should not
puase...between...the...words. Plaintalk is trained on recordings
of people calmly reading out lists of phrases, so it is best at
recognizing speech that is similarly spoken.
I forwarded your message directly to them -- to inspire them a bit by
getting a communication from someone who is responsible for the
technology. Today, we're taking a quiz. We'll see how they do.
#4. When training the Speech Recognition in the system preferences,
you should speak louder and faster until the test phrase blinks to
show that the phrase was recognized. True or False?
Do not post admin requests to the list. They will be ignored.
Speech-dev mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden