Method and system for generating a speech recognition dictionary based on greeting recordings in a voice messaging system

A method for generating a speech recognition dictionary, comprising:

• a) receiving a signal derived from audio greetings recorded by telephone system subscribers that are audibly played by the telephone system prior to allowing callers to leave voice messages to the telephone system subscribers, each greeting being indicative of a name of a subscriber of the telephone system;
• and
• b) processing the signal to generate a speech recognition dictionary, said speech recognition dictionary including vocabulary items representative of the names of the subscribers indicative by the greetings, said speech recognition dictionary being suitable for processing by a speech recognition system attempting to match a spoken utterance against a vocabulary item of said speech recognition dictionary.

A system to generate a speech recognition dictionary, said system comprising:

• a) input means for receiving a signal derived from audio greetings recorded by telephone system subscribers that are audibly played by the telephone system prior to allowing callers to leave voice messages to the telephone system subscribers, each greeting being indicative of a name of a subscriber of the telephone system;
• and
• b) processing means for processing the signal to generate a speech recognition dictionary, said speech recognition dictionary including vocabulary items representative of the names of the subscribers indicative by the greetings, said speech recognition dictionary being suitable for processing by a speech recognition system attempting to match a spoken utterance against a vocabulary item of said speech recognition dictionary.

A telephone system with voice messaging capability, comprising:

• a) a machine readable storage medium for storing greetings, each greeting being associated with a voice mailbox of a subscriber and being played prior to allowing a caller to a leave a voice messages to the telephone system subscriber;
• b) a processing unit coupled to said machine readable storage medium for processing the signal to generate a speech recognition dictionary, said speech recognition dictionary including vocabulary items representative of the names of the subscribers indicative by the greetings, said speech recognition dictionary being suitable for processing by a speech recognition system attempting to match a spoken utterance against a vocabulary item of said speech recognition dictionary;
• c) a voice messaging unit coupled to said machine readable storage medium operative for audibly playing a greeting when a condition exists indicative of an attempt of a caller to leave a message in a voice mail box of a subscriber.