Directory assistance system and method utilizing a speech recognition system and a live operator

• US 6,243,684 B1
• Filed: 02/19/1999
• Issued: 06/05/2001
• Est. Priority Date: 02/19/1999
• Status: Expired due to Term

• Current Assignee: Usada Incorporated
• Original Assignee: Usada Incorporated
• Inventors: Stuart, Robert O., Stuart, Scott P., Hornback, Don
• Primary Examiner(s): Dorvil, Richemond
• Application Number:
  US09/255,624
• Time in Patent Office: 837 Days
• Field of Search: 379/88.01, 704/275, 704/270, 704/272, 704/200, 704/231, 704/246, 704/251, 704/255
• US Class Current: 704/275
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 2203/357 Autocues for dialog assistance
• CPC Class Codes: H04M 3/4931 Directory assistance systems

Abstract

A method and apparatus utilizes a personal speech recognition system to enable directory assistance operators to retrieve customer requested information from a listing database based upon a voiced inquiry by the operator. Instead of requiring an operator to interpret a calling customer'"'"'s request and communicate the same to a computer via a number of keyboard strokes, the live operator may submit the same or equivalent inquiry via a voice utterance that is recognized by the searching computer as a searchable inquiry by a speech recognition system, which is preferably trained to recognize the particular voice of the live operator. Only the voice of the live operator needs to be identified by the voice recognition unit, and the incorporation of the voice recognition technology into the system is thus transparent to the calling customer. Once the appropriate listing is found, the call is released to the customer in a conventional manner, automatically, or after a particular voice command is uttered by the live operator to the computer.

Claims

1. A directory assistance system for providing requested information to a caller based upon identifying information provided by the caller, the system comprising:

means for conveying said identifying information to a live operator;

a computer having access to a directory database, and means for searching said database in response to an inquiry; and

a voice communication interface from said live operator to said computer that includes a speech recognition system capable of converting a spoken inquiry from said live operator into an interpreted searchable inquiry that is available to said means for searching.

2. The directory assistance system of claim 1 wherein said means for conveying includes an audio recording apparatus and a playback device.

3. The directory assistance system of claim 1 wherein said speech recognition system is specifically trained to recognize said live operator'"'"'s voice.

4. The directory assistance system of claim 1 further comprising a manual communication interface from said live operator to said computer.

5. The directory assistance system of claim 1 further comprising a direct voice communication interface from said caller to said computer that includes an independent speech recognition system capable of converting said identifying information spoken by said caller into an independent searchable inquiry that is available to said means for searching.

6. The directory assistance system of claim 5 further comprising means for merging a first search result based upon said independent searchable inquiry with a second search result based upon said interpreted searchable inquiry.

7. The directory assistance system of claim 1 further comprising a video display interface between said computer and said live operator that is capable of visually communicating a search result to said live operator.

8. A method of directory assistance for providing requested information to a caller based upon identifying information provided by the caller, comprising the steps of:

establishing a telephonic connection between a directory assistance system and a caller;

conveying said identifying information to a live operator;

positioning a speech recognition system between said live operator and a computer having access to a directory database;

searching said directory database at least partially in response to an utterance spoken by said live operator to said computer via said speech recognition system, wherein at least a portion of said utterance is an interpretation of at least a portion of said identifying information; and

releasing a search result to said caller.

9. The method of directory assistance of claim 8 wherein said conveying step includes the steps of:

recording a caller utterance made by said caller;

playing back a portion of said caller utterance to said live operator.

10. The method of directory assistance of claim 8 further comprising a step of specifically training said voice recognition system to recognize speech by said live operator.

11. The method of telephone directory assistance of claim 8 further comprising a step of identifying a group of most frequently requested listings that is a subset of said directory database.

12. The method of directory assistance of claim 11 further comprising the steps of:

determining whether said identifying information corresponds to one of said most frequently requested listings; and

if said identifying information corresponds to one of said most frequently requested listings, then searching said subset of said directory database at least partially in response to said utterance spoken by said live operator to said computer.

13. The method of directory assistance of claim 8 further comprising the steps of:

displaying a search result on a view screen visible to said live operator;

choosing a single listing if said search result includes a plurality of listings; and

releasing said single listing to said caller; and

at least one of said choosing step and said releasing step includes a voiced communication from said live operator to said computer.

14. The method of directory assistance system of claim 8 further comprising the steps of:

conveying said identifying information to a direct voice communication interface from said caller to said computer; and

converting said identifying information with an independent speech recognition system into an independent searchable inquiry.