Method and apparatus for performing speech recognition utilizing a supplementary lexicon of frequently used orthographies

A method for performing speech recognition, said method comprising the steps of:

• receiving a spoken utterance by a user of said speech recognition system;
• providing a speech recognition dictionary including a plurality of orthographies potentially recognizable on a basis of the spoken utterance;
• processing said speech recognition dictionary to derive on the basis of the spoken utterance a list of orthographies, said list containing a plurality of orthographies, each orthography in said list being a candidate having a certain probability to correspond to the spoken utterance;
• providing a standard text lexicon including a plurality of orthographies;
• inserting at least one orthography from said standard text lexicon into said list to form an augmented list.

A machine readable storage medium containing a program element for directing a computer to recognize a spoken utterance, the computer including:

• memory unit including:
• a) a speech recognition dictionary including a plurality of orthographies potentially recognizable on a basis of the spoken utterance;
• b) a standard text lexicon including a plurality of orthographies;
• processor in operative relationship with the memory unit, said program clement being executable by the processor and being operative for:
• a) processing said speech recognition dictionary to derive on the basis of the spoken utterance a list of orthographies, said list containing a plurality of orthographies, each orthography in said list being a candidate having a certain probability to correspond to the spoken utterance;
• b) inserting at least one orthography from the standard text lexicon into said list to form an augmented list.

A speech recognition apparatus comprising:

• means for receiving a spoken utterance by a user of said speech recognition apparatus;
• a speech recognition dictionary including a plurality of orthographies potentially recognizable on a basis of the spoken utterance;
• means for processing said speech recognition dictionary to derive on the basis of the spoken utterance a list of orthographies, said list containing a plurality of orthographies, each orthography in said list being a candidate having a certain probability to correspond to the spoken utterance;
• a standard text lexicon including a plurality of orthographies;
• means for inserting at least one orthography from said standard text lexicon into said list to form an augmented list.