Voice dialing using a rejection reference
First Claim
Patent Images
1. A voice dialing method, comprising the steps of:
- (a) receiving from a user multiple utterances each representative of at least one digit, using a microphone associated with an automatic speech recognition system;
(b) identifying multiple recognition results for the multiple utterances using a processor associated with the automatic speech recognition system and storing the recognition results at least one recognition result memory location associated with the automatic speech recognition system;
(c) communicating to the user the multiple recognition results from step (b) via an output device associated with the automatic speech recognition system;
(d) receiving from the user an indication that one of the multiple recognition results is incorrect; and
(e) populating, with the one incorrect recognition result, at least one digit position of a plurality of digit positions of a rejection reference memory location that is associated with the automatic speech recognition system and that is not the same as the recognition result memory location, wherein the rejection reference memory location is for cross-referencing against the recognition result memory location;
(f) receiving from the user another indication that another one of the multiple recognition results is incorrect;
(g) clearing the one recognition result from the rejection reference and populating the rejection reference with the another one of the recognition results after receiving the another indication;
(h) receiving a correction utterance from the user; and
then(i) processing the correction utterance based on the another one of the recognition results being in the rejection reference.
2 Assignments
0 Petitions
Accused Products
Abstract
A voice dialing method includes the steps of receiving an utterance from a user, decoding the utterance to identify a recognition result for the utterance, and communicating to the user the recognition result. If an indication is received from the user that the communicated recognition result is incorrect, then it is added to a rejection reference. Then, when the user repeats the misunderstood utterance, the rejection reference can be used to eliminate the incorrect recognition result as a potential subsequent recognition result. The method can be used for single or multiple digits or digit strings.
-
Citations
12 Claims
-
1. A voice dialing method, comprising the steps of:
-
(a) receiving from a user multiple utterances each representative of at least one digit, using a microphone associated with an automatic speech recognition system; (b) identifying multiple recognition results for the multiple utterances using a processor associated with the automatic speech recognition system and storing the recognition results at least one recognition result memory location associated with the automatic speech recognition system; (c) communicating to the user the multiple recognition results from step (b) via an output device associated with the automatic speech recognition system; (d) receiving from the user an indication that one of the multiple recognition results is incorrect; and (e) populating, with the one incorrect recognition result, at least one digit position of a plurality of digit positions of a rejection reference memory location that is associated with the automatic speech recognition system and that is not the same as the recognition result memory location, wherein the rejection reference memory location is for cross-referencing against the recognition result memory location; (f) receiving from the user another indication that another one of the multiple recognition results is incorrect; (g) clearing the one recognition result from the rejection reference and populating the rejection reference with the another one of the recognition results after receiving the another indication; (h) receiving a correction utterance from the user; and
then(i) processing the correction utterance based on the another one of the recognition results being in the rejection reference. - View Dependent Claims (2, 3, 4, 5, 6)
-
-
7. A voice dialing method, comprising the steps of:
-
(a) receiving multiple utterances from a user using a microphone associated with an automatic speech recognition system; (b) determining multiple recognition results for the multiple utterances using a processor associated with the automatic speech recognition system and storing the recognition results in at least one recognition result memory location associated with the automatic speech recognition system; (c) communicating the multiple recognition results to the user, using an output device associated with the automatic speech recognition system; (d) receiving from the user an indication of whether one of the multiple recognition results is incorrect, using a user interface associated with the automatic speech recognition system; and (e) if incorrect, then populating, with the one incorrect recognition result, a rejection reference memory location associated with the automatic speech recognition system wherein the rejection reference memory location is for cross-referencing against the recognition result memory location and includes a matrix having a plurality of rows corresponding to misrecognized utterances and a plurality of columns of digit positions corresponding to discrete digits of the misrecognized utterances; (f) receiving from the user another indication that another one of the multiple recognition results is incorrect; (g) clearing the one recognition result from the rejection reference and populating the rejection reference with the another one of the recognition results after receiving the another indication; (h) receiving a correction utterance from the user; and
then(i) processing the correction utterance based on the another one of the recognition results being in the rejection reference.
-
-
8. A voice dialing method, comprising the steps of:
-
(a) receiving an initial user utterance at an automatic speech recognition system having computer readable memory and a processor that access program instructions from the computer readable memory and executes the program instructions to carry out speech recognition, wherein the initial utterance corresponds to a number, which has a plurality of digits; (b) determining an initial recognition result for the initial user utterance using the processor of the automatic speech recognition system, and using an initial grammar to recognize the initial user utterance; (c) receiving a subsequent user utterance at the automatic speech recognition system; and (d) determining a subsequent recognition result for the subsequent user utterance, using the processor, and using a subsequent grammar to recognize the subsequent user utterance, wherein the initial grammar corresponds to an initial quantity of digits and the subsequent grammar corresponds to subsequent quantity of digits less than the initial quantity of digits. - View Dependent Claims (9, 10, 11, 12)
-
Specification