METHOD AND APPARATUS FOR GENERATING A SPEECH RECOGNITION DICTIONARY BASED ON GREETING RECORDINGS IN A VOICE MESSAGING SYSTEM

摘要

A system and a method for generating a speechrecognition dictionary that can be used in a telephone systemhaving speech recognition capabilities, in particularcapabilities to effect a connection when the calling partyutters the name cf a subscriber (called party). The methodgenerates transcriptions associated to respective vocabularyitems in the speech recognition dictionary from audiogreetings recorded by .the telephone system subscribers.Normally such audio greetings are used in voice messagingapplications. Typically, the greetings are played beforeallowing callers to leave messages in a voice mailbox ofsubscribers. An individual greeting is audio information thatcontains the name of the subscriber. This audio informationis processed to generate a transcription associated to avocabulary item in the speech recognition dictionary,representative of the subscriber name.