Appeal No. 2003-2018
Application No. 09/057,261
can be used to distinguish one sound from another. The sequence
of phoneme codes is decoded into a literal string of words using
a phonetic dictionary and a syntax file. See page 1 of
Appellant's specification. These conventional systems are not
suitable for use in communication systems using mobile
communication units. See page 3 of Appellant's specification.
Appellant's speech recognition system performs speech decoding at
the mobile terminal as compared to a host computer. Appellant's
system utilizes a mobile terminal employing virtual Graphical
User Interface (GUI) pages to facilitate user interface.
Associated with each GUI file is a Hypertext Markup Language
(HTML) file, dictionary file which includes corresponding speech
recognition identifying data (e.g., phoneme files) and a syntax
file. The dictionary files and syntax files are limited in scope
to provide for identification of commands associated primarily
only with the particular GUI file. This greatly reduces the
memory requirements because the dictionary file(s) and syntax
file(s) are limited to words associated with the limited commands
and data which the user of the mobile terminal could validly
input in connection with a particular GUI file. See page 4 of
Appellant's specification.
22
Page: Previous 1 2 3 4 5 6 7 8 9 10 Next
Last modified: November 3, 2007