Background speech recognition for voice messaging applications

• US 6,035,017 A
• Filed: 01/24/1997
• Issued: 03/07/2000
• Est. Priority Date: 01/24/1997
• Status: Expired due to Term

• Current Assignee: Avaya Incorporated
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Rabiner, Lawrence Richard, Pilc, Randolph John, Fenton, Francis Michael
• Primary Examiner(s): Hunter, Daniel S.
• Application Number:
  US08/788,752
• Time in Patent Office: 1,138 Days
• Field of Search: 704/251, 704/270, 704/275, 379/67, 379/88, 379/89, 379/142, 379/355, 379/201, 379/211, 379/88.01, 379/88.04, 379/88.13, 379/88.14, 379/88.15
• US Class Current: 379/88.04
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 2201/60 Medium conversion
• CPC Class Codes: H04M 3/533 Voice mail systems
• CPC Class Codes: H04M 3/5335 Message type or catagory, e.g. priority, indication

Abstract

Recorded telephone messages may be automatically handled as specified by the called party by means of a background speech recognition server for a telephone network, a private branch exchange, or a personal computer telephone manager. The called party initializes the method by storing multiple action records in a computer, each containing a key word and an action description. Incoming telephone messages can then be automatically processed by subjecting the audio record of each message to a speech recognition program to produce a text record. The computer then compares the text record with the key words in the action records. When a match is found, the action specified in the action record bearing the matching key words, is performed. In this manner, recorded telephone messages may be automatically handled and prioritized as specified by the called party.

Claims

1. A method for managing voice messages for use on a voice messaging system, the method comprising the steps of:

receiving from a user of the voice messaging system a spoken phrase;

creating an association between the spoken phrase and an action description, which is selected by the user; and

subsequently processing a voice message as a function of the created association by

(1) searching for a match between portions of the voice message and the spoken phrase, and

(2) if a match is found, performing the action, a priori associated with the action description, on the voice message.

2. The method of claim 1, wherein the creating step includes the steps of creating an action record comprising a name key field for representing name information of a calling party, a phone key field for representing information about a calling party number, a phrase key field for storing text representing the spoken phrase, and the action description.

3. The method of claim 1 wherein the searching step also searches for matches in the name key field and the call party number field for performing the associated action.

4. A method for managing voice messages for use on a voice messaging system, which provides a voice messaging service to a number of users, the method comprising the steps of:

enabling each user to edit their own action records in the voice messaging system, each action record comprising a representation of a spoken phrase, provided by the user, and an action description, selected by the user, such that each user can define their own association between a representation of a spoken phrase and an action description independent of other user'"'"'s defined action records;

subsequently processing received voice messages for the users as a function of their respective action records by;

(1) searching for a match between portions of each received voice message for each user and the corresponding portions of each user'"'"'s action records representing spoken phrases; and

(2) if a match is found for one of the user'"'"'s action records, performing an action, a priori associated with the action description, on the corresponding voice message.

5. A method for managing voice messages for use on a voice messaging system, which provides a voice messaging service to a number of users, the method comprising the steps of:

enabling each user to edit their own action records in the voice messaging system, each action record comprising a name key field for representing name information of a calling party, a phone key field for representing information about a calling party number, a phrase key field for storing text representing a spoken phrase, and an action description, wherein contents of the name key field, phone key field, and phrase key field are created by each user and the action description is selected, from one of a number of selections provided by the voice messaging system, by each user such that each user can define their own association between the key fields and a selected action description independent of other user'"'"'s defined action records;

subsequently processing received voice messages for the users as a function of their respective action records by;

(1) searching for matches between calling party name, calling party number, and spoken phrases of each received voice message for each user and the corresponding key fields of each user'"'"'s action records; and

(2) if a match is found for one of the user'"'"'s action records, performing an action, a priori associated with the action description of the matched action record, on the corresponding voice message.

6. A method for managing voice messages for use on a voice messaging system, which provides a voice messaging service to a number of users, the method comprising the steps of:

enabling each user to edit their own action records in the voice messaging system, each action record comprising a name key field for representing name information of a calling party, a phone key field for representing information about a calling party number, a phrase key field for storing text representing a spoken phrase, and an action description, wherein contents of the name key field, phone key field, and phrase key field are created by each user and the action description is selected, from one of a number of selections provided by the voice messaging system, by each user such that each user can define their own association between the key fields and a selected action description independent of other user'"'"'s defined action records;

subsequently processing each user'"'"'s received voice messages as a function of their respective action records by;

(1) searching for a match between a calling party name of each received voice message, and contents of the name key field of each user'"'"'s action records;

(2) if a match is found for one of the user'"'"'s action records, checking a quality level of the match;

(3) if the checked quality level is above a predetermined threshold, performing an action, a priori associated with the action description of the matched action record, on the corresponding voice message;

(4) if the checked quality level is not above a predetermined threshold, comparing content of the phone key field of the matched action record with a calling party number of the corresponding received voice message; and

(5) if the content of the phone key field matches the calling party number, performing the action, a priori associated with the action description of the matched action record, on the corresponding voice message.

7. The method of claim 6 further comprising the steps of:

(6) if there is no match of the name key field, searching for a match between portions of the received voice message and content of the phrase key field; and

(7) if a match is found, performing the action, a priori associated with the action description of the matched action record, on the corresponding voice message.