Apparatus and method for identifying prosody and apparatus and method for recognizing speech
First Claim
Patent Images
1. A prosody identifying apparatus for identifying input speech on the basis of prosodic features of the input speech, comprising:
- sound analyzing means for acquiring an amount of change in movement of a feature distribution obtained from an autocorrelation matrix of the frequency characteristic of the input speech; and
identifying means for recognizing the input speech on the basis of an output of the sound analyzing means.
1 Assignment
0 Petitions
Accused Products
Abstract
A prosody identifying apparatus for identifying input speech on the basis of prosodic features of the input speech is provided. The prosody identifying apparatus includes a sound analyzing section for acquiring an amount of change in movement of a feature distribution obtained from an autocorrelation matrix of the frequency characteristic of the input speech and an identifying section for recognizing the input speech on the basis of an output of the sound analyzing section.
-
Citations
10 Claims
-
1. A prosody identifying apparatus for identifying input speech on the basis of prosodic features of the input speech, comprising:
-
sound analyzing means for acquiring an amount of change in movement of a feature distribution obtained from an autocorrelation matrix of the frequency characteristic of the input speech; and identifying means for recognizing the input speech on the basis of an output of the sound analyzing means. - View Dependent Claims (2, 3)
-
-
4. A method for identifying input speech on the basis of prosodic features of the input speech, comprising the steps of:
-
acquiring an amount of change in movement of a feature distribution obtained from an autocorrelation matrix of the frequency characteristic of the input speech; and recognizing the input speech on the basis of an output of the step of acquiring an amount of change in movement of a feature distribution.
-
-
5. A speech recognition apparatus comprising:
-
inputting means for inputting a speech signal; prosody identifying means for identifying prosody on the basis of an amount of change in movement of a feature distribution obtained from an autocorrelation matrix of a frequency characteristic of the input speech input to the inputting means; and speech recognizing means for performing speech recognition on the basis of features acquired by sound-analyzing the speech input to the inputting means; and selecting means for selecting whether to output an output of the speech recognizing means or to integrate the output from the prosody identifying means with an output of the speech recognizing means and output the integrated output, the selecting means outputting the selected output. - View Dependent Claims (6, 7)
-
-
8. A method for recognizing input speech, comprising the steps of:
-
inputting a speech signal; identifying prosody on the basis of an amount of change in movement of a feature distribution obtained from an autocorrelation matrix of a frequency characteristic of the input speech input in the step of inputting a speech signal; and performing speech recognition on the basis of features acquired by sound-analyzing the speech input in the step of inputting a speech signal; and selecting whether to output an output of the step of identifying prosody or to integrate the output from the step of identifying prosody with an output from the step of performing speech recognition and output the integrated output, and outputting the selected output.
-
-
9. A prosody identifying apparatus for identifying input speech on the basis of prosodic features of the input speech, comprising:
-
a sound analyzing section configured to acquire an amount of change in movement of a feature distribution obtained from an autocorrelation matrix of the frequency characteristic of the input speech; and an identifying section configured to recognize the input speech on the basis of an output of the sound analyzing section.
-
-
10. A speech recognition apparatus comprising:
-
an inputting unit configured to input a speech signal; a prosody identifying unit configured to identify prosody on the basis of an amount of change in movement of a feature distribution obtained from an autocorrelation matrix of a frequency characteristic of the input speech input to the inputting unit; and a speech recognizing unit configured to perform speech recognition on the basis of features acquired by sound-analyzing the speech input to the inputting unit; and a selecting unit configured to select whether to output an output of the speech recognizing unit or to integrate the output from the prosody identifying unit with an output of the speech recognizing unit and output the integrated output, the selecting unit outputting the selected output.
-
Specification