Method and apparatus handling speech recognition errors in spoken dialogue systems
First Claim
1. An apparatus handling speech recognition errors in a spoken dialogue system, comprising:
- a speech recognition unit extracting a speech feature of a speech signal inputted from a user and finding a phoneme nearest to the speech feature to recognize a word;
a recognition error determination unit finding a sentence confidence based on a confidence of the recognized word, inspecting a semantic structure of a recognized sentence, and determining whether an error exists in the recognized sentence which is subject to speech recognition according to a predetermined criterion based on both the sentence confidence and a result of inspecting the semantic structure of the recognized sentence; and
a meta-dialogue generation unit generating a question questioning the user about the recognized sentence based on a content of a portion where the error exists and a type of the error.
1 Assignment
0 Petitions
Accused Products
Abstract
In order to handle portions of a recognized sentence having an error, a speaker or user is questioned about contents associated with the portions, and according to a user'"'"'s answer a result is obtained. A speech recognition unit extracts a speech feature of a speech signal inputted from a user and finds a phoneme nearest to the speech feature to recognize a word. A recognition error determination unit finds a sentence confidence based on a confidence of the recognized word, performs examination of a semantic structure of a recognized sentence, and determines whether or not an error exists in the recognized sentence which is subject to speech recognition according to a predetermined criterion based on both the sentence confidence and a result of examining the semantic structure of the recognized sentence. Further, a meta-dialogue generation unit generates a question asking the user for additional information based on a content of a portion where the error exists and a type of the error.
169 Citations
34 Claims
-
1. An apparatus handling speech recognition errors in a spoken dialogue system, comprising:
-
a speech recognition unit extracting a speech feature of a speech signal inputted from a user and finding a phoneme nearest to the speech feature to recognize a word;
a recognition error determination unit finding a sentence confidence based on a confidence of the recognized word, inspecting a semantic structure of a recognized sentence, and determining whether an error exists in the recognized sentence which is subject to speech recognition according to a predetermined criterion based on both the sentence confidence and a result of inspecting the semantic structure of the recognized sentence; and
a meta-dialogue generation unit generating a question questioning the user about the recognized sentence based on a content of a portion where the error exists and a type of the error. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
-
-
16. A method of handling speech recognition errors in a spoken dialogue system, comprising:
-
extracting a speech feature of a speech signal inputted by a user and finding a phoneme nearest to the speech feature to recognize a word;
determining whether an error exists in a sentence which is subject to speech recognition according to a predetermined criterion based on both a sentence confidence obtained from a confidence of the recognized word and a result of examining a semantic structure of the sentence; and
generating a question questioning the user about the recognized sentence based on both a content of a portion where the error exists and a type of the error. - View Dependent Claims (17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29)
-
-
30. An apparatus handling speech recognition errors, comprising:
-
a speech signal input unit receiving a speech signal input from a user;
a speech recognition unit extracting a speech feature from the speech signal input from the user and recognizing words, generating a plurality of candidate words and forming a recognized sentence;
a recognition error determination unit determining whether there is an error in the recognized sentence according to a predetermined criterion based on sentence confidence and semantic structure of the recognized sentence; and
a meta-dialogue generation unit generating a question questioning the user about the recognized sentence based on both a content of a portion where the error exists and a type of the error. - View Dependent Claims (31, 32)
-
-
33. A computer readable medium encoded with processing instructions performing a method of handling speech recognition errors in a spoken dialogue system, the method comprising:
-
extracting a speech feature of a speech signal inputted by a user and finding a phoneme nearest to the speech feature to recognize a word and generating a recognized sentence;
determining whether an error exists in the recognized sentence according to a predetermined criterion based on both a sentence confidence obtained from a confidence of the recognized word and a result of examining a semantic structure of the recognized sentence; and
generating a question questioning the user about the recognized sentence based on both a content of a portion where the error exists and a type of the error. - View Dependent Claims (34)
-
Specification