Method and apparatus for recognizing speech information based on prediction
First Claim
Patent Images
1. An information processing apparatus comprising:
- storage means for storing prediction information predicting a next phoneme, syllable, or word to be input based upon the meaning of one or more previously input words;
first determining means for determining whether inputted sound information is produced by a human;
means for acquiring information concerning specific language or non-language human sounds;
second determining means for determining whether the inputted sound information is the specific language or non-language human sounds;
third determining means for determining that the inputted sound information is language information in the event that said first determining means determines that the inputted sound information is produced by a human and said second determining means determines that the inputted sound information is not the specific language or non-language human sounds;
recognition means for recognizing the inputted sound information as a specific word, syllable, or phoneme, based on the prediction information predicting the current input word, syllable, or phoneme as a next word, syllable, or phoneme to be input based upon the meaning of one or more previously input words in the event that said third determining means determines that the inputted sound information is language information;
a knowledge base for storing knowledge concerning a type of data represented by the sound information and specific language or non-language human sounds;
prediction means for predicting a next word, syllable or phoneme to be input based upon the meaning of one or more previously input words by referring to the knowledge stored in said knowledge base; and
updating means for updating the prediction information stored in said storage means based on a prediction result obtained by said prediction means.
1 Assignment
0 Petitions
Accused Products
Abstract
An apparatus for recognizing sound information includes a sound recognition unit for recognizing sound information. A knowledge base stores knowledge concerning a type of data represented by the sound information. A prediction unit predicts the type of data represented by sound information to be subsequently recognized by checking the knowledge stored in the knowledge base. The sound recognition unit recognizes sound information to be subsequently recognized based on a prediction result obtained by the prediction unit. With this arrangement, the recognition accuracy is enhanced.
68 Citations
77 Claims
-
1. An information processing apparatus comprising:
-
storage means for storing prediction information predicting a next phoneme, syllable, or word to be input based upon the meaning of one or more previously input words;
first determining means for determining whether inputted sound information is produced by a human;
means for acquiring information concerning specific language or non-language human sounds;
second determining means for determining whether the inputted sound information is the specific language or non-language human sounds;
third determining means for determining that the inputted sound information is language information in the event that said first determining means determines that the inputted sound information is produced by a human and said second determining means determines that the inputted sound information is not the specific language or non-language human sounds;
recognition means for recognizing the inputted sound information as a specific word, syllable, or phoneme, based on the prediction information predicting the current input word, syllable, or phoneme as a next word, syllable, or phoneme to be input based upon the meaning of one or more previously input words in the event that said third determining means determines that the inputted sound information is language information;
a knowledge base for storing knowledge concerning a type of data represented by the sound information and specific language or non-language human sounds;
prediction means for predicting a next word, syllable or phoneme to be input based upon the meaning of one or more previously input words by referring to the knowledge stored in said knowledge base; and
updating means for updating the prediction information stored in said storage means based on a prediction result obtained by said prediction means. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 74, 76)
-
-
37. An information processing method comprising:
-
a first determining step for determining whether inputted sound information is produced by a human;
an acquiring step for acquiring information concerning specific language or non-language human sounds;
a second determining step for determining whether the inputted sound information is the specific language or non-language human sounds;
third determining step for determining that the inputted sound information is language information in the event that said first determining step determines that the inputted sound information is produced by a human and said second determining step determines that the inputted sound information is not the specific language or non-language human sounds;
a recognition step of recognizing the inputted sound information to be a specific word, syllable, or phoneme based on prediction information predicting the current input word, syllable, or phoneme as a next word, syllable, or phoneme to be input based upon the meaning of one or more previously input words in the event that inputted sound information is determined by said third determining step to be language information;
a prediction step of predicting a next word, syllable, or phoneme to be input based upon the meaning of one or more previously input words by checking knowledge stored in a knowledge base for storing knowledge concerning a type of data represented by sound information and for storing the specific language or non-language human sounds; and
an updating step of updating the prediction information based on a prediction result obtained in said prediction step. - View Dependent Claims (38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 75, 77)
-
-
73. A computer-readable storage medium storing a response program for controlling a computer to perform speech recognition, said program comprising codes for causing the computer to perform:
-
a first determining step for determining whether inputted sound information is produced by a human;
an acquiring step for acquiring information concerning specific language or non-language human sounds;
a second determining means for determining whether the inputted sound information is the specific language or non-language human sounds;
third determining step for determining that the inputted sound information is language information in the event that said first determining step determines that the inputted sound information is produced by a human and said second determining step determines that the inputted sound information is not the specific language or non-language human sounds;
a recognition step of recognizing the inputted sound information as a specific word, syllable, or phoneme based on prediction information predicting the current input word, syllable, or phoneme as a next input word, syllable, or phoneme to be input based upon the meaning of one or more previously input words in the event that inputted sound information is determined by said third determining step to be language information;
a prediction step of predicting a next word, syllable, or phoneme to be input based upon the meaning of one or more previously input words by checking knowledge stored in a knowledge base for storing knowledge concerning a type of data represented by sound information and for storing the specific language or non-language human sounds; and
an updating step of updating the prediction information based on a prediction result obtained in said prediction step.
-
Specification