A method and apparatus for improving the utilityof speech recognition is described. The method involvescapturing a spoken word, passing the spoken word to a speechrecognition algorithm, receiving at least one textrepresentation of the spoken word from the speechrecognition algorithm, and passing a text representation ofthe spoken word to a display telephone to permit the user toselect the correct representation of the voice response.The apparatus is an access server which communicates with adisplay telephone, a speech recognition algorithm whichresponds to queries from the access server and one or moredatabases which likewise respond to queries from the accessserver. The method and apparatus are particularly useful inautomating such functions as telephone directory servicesusing display telephones. The advantage is the ability tocompletely automate directory services for owners of displaytelephones and to significantly broaden the applications forspeech recognition as a tool in information retrieval andtransaction processing.
展开▼