Disambiguation of alphabetic characters in an automated call processing environment

• US 5,917,890 A
• Filed: 12/29/1995
• Issued: 06/29/1999
• Est. Priority Date: 12/29/1995
• Status: Expired due to Term

• Current Assignee: AT&T Corporation (AT&T, Inc.)
• Original Assignee: AT&T Corporation (AT&T, Inc.)
• Inventors: Farber, James M., Brotman, Lynne Shapiro, Stern, Benjamin J.
• Primary Examiner(s): WEAVER, SCOTT LOUIS
• Application Number:
  US08/581,716
• Time in Patent Office: 1,278 Days
• Field of Search: 379/67, 379/77, 379/88, 379/89, 379/201, 379/213, 379/354, 379/355, 379/67.1, 379/88.01, 395/2, 395/2.4, 395/2.6, 395/2.61, 395/2.62, 395/2.63, 395/2.64, 395/2.66, 395/2.79, 395/2.84, 704/231, 704/251, 704/252, 704/253, 704/254, 704/255, 704/270, 704/275
• US Class Current: 379/88.01
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: H04M 1/271 controlled by voice recognition
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/42 Systems providing special services or facilities to subscribers specially adapted for wireless communication networks H04W4/00
• CPC Class Codes: H04M 3/493 Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
• CPC Class Codes: H04M 3/533 Voice mail systems

Abstract

Automated capture of an uttered alphabetic character is provided by using an input, beyond the uttered alphabetic character, to disambiguate an incorrectly captured character. The input is an indication of a telephone key representing the uttered alphabetic character. The indication can be a dual tone multifrequency signal or an utterance of the number of the telephone key. Alternatively, the input is an indication that the incorrectly captured alphabetic character differs from the uttered alphabetic character.

Claims

1. A method of capturing at least one uttered alphabetic character, comprising the steps of:

receiving a signal indicative of an uttered alphabetic character;

accessing a set of stored signals representing spoken alphabetic characters;

automatically finding a first candidate alphabetic character from the set of stored signals corresponding to the received signal;

inquiring from a user whether the first candidate alphabetic character is the uttered alphabetic character; and

receiving an input for use in disambiguating the received signal when the first candidate alphabetic character differs from the uttered alphabetic character, wherein the input is an indication of a telephone key representing the uttered alphabetic character and wherein the telephone key indication is an utterance; and

disambiguating the received signal, by automatically finding an alternate candidate alphabetic character that accurately captures the uttered character;

wherein the step of automatically finding the alternate candidate alphabetic character includes using the input to compare the stored signals with the received signal until one of the stored signals matches the received signal.

2. A method of accurately capturing at least one uttered alphabetic character, comprising the steps of:

receiving a signal indicative of an uttered alphabetic character;

accessing a set of stored signals representing spoken alphabetic characters;

automatically finding a first candidate alphabetic character corresponding to the received signal;

inquiring from a user whether the first candidate alphabetic character is the uttered alphabetic character;

receiving an input for use in disambiguating the received signal when the first candidate alphabetic character differs from the uttered alphabetic character;

automatically finding a second candidate alphabetic character corresponding to the received signal in accordance with the input, wherein the step of automatically finding the second candidate alphabetic character includes comparing the stored signals with the received signal;

eliminating a stored signal representing the first candidate alphabetic character from the stored signals and inquiring whether the second candidate alphabetic candidate character accurately captures the uttered character;

wherein

automatically finding an alternative candidate alphabetic character;

wherein the step of automatically finding includes using the input to compare a selected group of the stored signals representing the spoken alphabetic characters indicated by the input with the received signal until one of the stored signals matches the received signal.

3. Apparatus for capturing at least one uttered alphabetic character, comprising:

means for receiving a signal indicative of an uttered alphabetic character;

means for a set of stored signals representing spoken alphabetic characters;

means for automatically finding a first candidate alphabetic character corresponding to the received signal;

means for inquiring from a user whether the first candidate alphabetic character is the uttered alphabetic character;

means for receiving an input for use in disambiguating the received signal when the first candidate alphabetic character differs from the uttered alphabetic character, wherein the input is an indication of a telephone key representing the uttered alphabetic character and wherein the telephone key indication is an utterance;

disambiguating the received signal, by automatically finding an alternate candidate alphabetic character that accurately captures the uttered character;

wherein the step of automatically finding the alternate candidate alphabetic character includes using the input to compare a selected group of the stored signals with the received signal until one of the stored signals matches the received signals.

4. The apparatus of claim 3, wherein the utterance represents a number corresponding to the telephone key.

5. Apparatus for capturing at least one uttered alphabetic character, comprising:

means for receiving a signal indicative of an uttered alphabetic character;

means for automatically finding a first candidate alphabetic character corresponding to the received signal;

means for accessing a set of stored signals representing spoken alphabetic characters;

means for inquiring from a user whether the first candidate alphabetic character is the uttered alphabetic character;

means for receiving an input for use in disambiguating the received signal when the first candidate alphabetic character differs from the uttered alphabetic character; and

means for automatically finding a second candidate alphabetic character corresponding to the received signal in accordance with the input;

wherein the means for automatically finding the second candidate alphabetic character includes means for comparing the stored signals with the received signal and wherein a stored signal representing the first candidate alphabetic character is eliminated from the stored signals indicated by the input;

disambiguating the received signal, by automatically finding an alternate candidate alphabetic character that accurately captures the uttered character;

wherein the step of automatically finding the alternate candidate alphabetic character includes the step of using the input to compare a selected group of the stored signals representing the first spoken alphabetic characters indicated by the input with the received signal until one of the stored signals matches the received signal.